Saltroad, the clinician-led speech and language therapy (SLT) provider, has raised £1.5 million and acquired Ogma, an AI documentation platform designed for SLT.  The round was led by Techstart Ventures, with participation from Ascension, ScaleX and a group of prominent angel investors. An estimated 1 in 5 children - around two million in the UK - need support with speech, language and communication, and demand far outstrips supply. Services built for a different era can't keep pace, and families and therapists have been left to absorb the trade-offs: reports instead of therapy, rationed time, endless admin and caseloads that don’t fit. Saltroad's premise is that none of this is inevitable.  It gives families private access to NHS speech and language therapists, meaning a wider pool of specialists to choose from, therapists matched to their child's clinical needs, no waiting lists, and support that costs less than traditional private therapy. Integrating the Ogma acquisition turns the raw material of a therapy session into structured, clinically useful notes - cutting after-hours admin and standardising record quality across the workforce.  By embedding AI tooling across its network of over 1,000 therapists across the UK, Saltroad aims to increase the number of children each therapist can see without compromising clinical quality or therapist wellbeing. According to Darren Lester, co-founder and CEO of Saltroad, too many children wait months, sometimes years, for help during the years that matter most.  “That isn't a failure of effort from therapists - it's a system that was never built for the scale or variety of need it now faces. Saltroad exists to put the therapy back into speech and language therapy, and to reach the families the current systems can't. This funding, and the Ogma acquisition, supports how we build that and deliver tailored, 1:1 speech therapy at scale.” Audrey Osborne, Partner at Techstart Ventures, said: "We've backed Saltroad since inception, and everything since has deepened our conviction in Darren, Debi and the team. The combination of a scalable therapist workforce and purpose-built AI is exactly the kind of ambition we want to support in Northern Ireland." Toyosi Ogedengbe, Partner at Ascension, said: "We look for teams solving problems that genuinely matter, at scale, and Saltroad does exactly that. From the start, Saltroad has been clear-eyed about a real and growing problem. Bringing AI alongside skilled clinicians, rather than in place of them, is the right way to widen access without lowering the bar." The funding and acquisition support Saltroad's plan to build an AI-enabled SLT workforce - pairing the flexibility of an associate model with purpose-built tools that free therapists to spend more time with the children who need support, and less on admin.

From June 17 to 20, 2026, VivaTech welcomed more than 200,000 visitors at Paris Porte de Versailles, setting a new attendance record for its 10th anniversary edition. Held under the joint presence of French President Emmanuel Macron and Indian Prime Minister Narendra Modi, the show confirmed its position as one of Europe's leading events dedicated to innovation, tech, and startups, and as one of the most influential tech events in the world. A decade of scale: the numbers behind the milestone The 2026 edition stretched VivaTech's footprint further than ever before. Over four days, the event brought together 200,000 visitors from 165 nationalities, hosted 60 country pavilions, and gathered more than 1,155 speakers on stage. More than 15,000 startups participated, alongside over 4,500 exhibitors, 61% of them international. The cumulative social media audience surpassed 5 billion, reinforcing VivaTech's role as a global conversation driver around innovation. To mark its decade in style, VivaTech also took the experience outside the venue: on Sunday, June 14, in partnership with the Comité Champs-Élysées, the world's most famous avenue was transformed into a large open-air technology showcase, opening innovation to the wider public ahead of the trade days. A program built around the defining themes of the decade ahead This anniversary edition was structured around five strategic themes shaping the next wave of technological transformation: AI & Productivity, Cybersecurity & Defense, GreenTech, Space, and DeepTech. The speaker line-up reflected the breadth of these topics. Among them: Jeff Bezos, Founder of Amazon & Blue Origin and Co-CEO of Prometheus; Dave Limp, CEO of Blue Origin; former NASA astronaut Mike Massimino; Yann LeCun, Executive Chairman of AMI Labs; Shantanu Narayen, Chair & CEO of Adobe; Sir Tim Berners-Lee, inventor of the World Wide Web; and Thibault Sottiaux, Head of Core Product & Platform at OpenAI. European institutions were also strongly represented, with Henna Virkkunen, Executive Vice-President for Tech Sovereignty, Security and Democracy at the European Commission, and Ekaterina Zaharieva, EU Commissioner for Startups, Research and Innovation. French and international business leaders, including Bernard Arnault (LVMH), Patrick Pouyanné (TotalEnergies), Christel Heydemann (Orange), Catherine MacGregor (ENGIE), Marie-Ange Debon (Groupe La Poste), Rodolphe Saadé (CMA CGM), Roland Busch (Siemens), Joe Tsai (Alibaba), Valérie Baudson (Amundi) and Mohamed Kande (PwC), joined the conversation. Germany was named Country of the Year 2026, with a strong ministerial delegation including Karsten Wildberger, Federal Minister for Digital Transformation and Government Modernization, and Dorothée Bär, Federal Minister for Research, Technology, and Space. Building on the AI Summit in New Delhi, India was present as AI Country Partner 2026, led by Prime Minister Narendra Modi. From conversations to business: new formats for acceleration True to its DNA, VivaTech 2026 doubled down on its business utility. New dedicated spaces were introduced, including the Business Plaza, designed to turn connections into concrete commercial opportunities, and Investors Office Hours, a privileged meeting format between startups and investors. With more than 15,000 startups connecting to companies from over 30 sectors, the event continued to position itself as a deal-making hub for the European and global tech ecosystem. New flagship exhibitors joined the show for the first time, including Samsung, Envision, SAP, and Adobe. Innovations that defined the edition VivaTech 2026 served as a launchpad for hundreds of announcements and product premieres. Standout innovations included XPANCEO's smart contact lens, designed to replace screens as the leading interface of the AI era; Lifepods' autonomous protection capsule for extreme-risk situations; Lattice Medical's 3D-printed resorbable implants for soft tissue reconstruction after cancer treatment; Tetmet's automated robotic solution for producing stainless steel mesh for the automotive, defense and aerospace industries; and a striking demonstration by Unitree x HABS of a humanoid robot controlled by human brain activity. Honoring tech's leaders: the VivaTech x Bloomberg Awards For the first time, the VivaTech x Bloomberg Awards recognized the world's top tech leaders. The Visionary Award went to Sir Tim Berners-Lee (Inrupt), the Leadership Award to Joe Tsai (Alibaba), the Momentum Award to Yann LeCun (AMI Labs), the Breakthrough Award to Peter Steinberger (OpenClaw / OpenAI), the Investor Award to Jeannette zu Fürstenberg (General Catalyst), the Rising Star Award to May Habib (WRITER), and the CitizenTech Award to Ukraine. The VivaTech Startup Prizes spotlighted impact-driven entrepreneurs: Liz Dennett (Endolith) received the Female Founder Award, Fanny Giannou (Alithea Biotechnology) the Tech for Change Award, Ahmed Yahia (Surgia) the AfricaTech Award, Sasha Ovalle (AssisTech Smart Shower) the Next Startupper Challenge, and Karim Boussetta (Hodor) the Innovation of the Year Award. Opening the next decade On Saturday, June 20, VivaTech opened its doors to the general public for a day dedicated to AI, robotics and career encounters, with French ESA astronaut Thomas Pesquet as special guest. As underlined by Maurice Lévy, Co-President of VivaTech and Emeritus Chairman of Publicis Groupe, Michèle Benbunan, Co-President of VivaTech and CEO of LVMH's Press division, and François Bitouzet, Managing Director of VivaTech, this 10th edition was not designed as a celebration of past years, but as the opening of a new decade of innovation. Next stop: VivaTech 2027, from June 16 to 19, at Paris Expo Porte de Versailles.

Lucida AI, a speech-to-speech AI platform for global communication, has secured an additional $1.6 million in funding, bringing its seed round to $7 million. The seed round was led by UK-based Velocity Capital, with participation from Next Tier Ventures, Look AI Ventures, Bogazici Ventures, Yapı Kredi Frwrd Ventures and Ünlü & Co. The company had announced the initial $5.4 million close of its seed round a few months ago. As communication increasingly spans multiple languages and cultures, spoken fluency and confidence play a growing role in how effectively people interact. Founded by Mustafa Girgin and M. Sait Demirci, Lucida AI develops speech-native AI technology designed to help individuals and businesses improve spoken communication through real-time, AI-powered conversations. Powered by its proprietary Speech Language Model (SLM), the platform enables users to speak naturally with AI without relying on text prompts or scripted exercises. It adapts conversations to each user's proficiency level, providing instant feedback on fluency, pronunciation and clarity while simulating real-world situations ranging from everyday interactions to business meetings, presentations and client calls. Lucida AI is available as a mobile application for individual users and also offers enterprise deployments with on-premises hosting and end-to-end encryption. Since launching 15 months ago, the company says it has grown to more than 3 million users and generated over 2.2 billion minutes of spoken interaction across Europe, the United States and emerging markets. AI is becoming a global race, and meaningful innovation is no longer tied to a single geography. With this round, we've partnered with investors who share our long-term vision. Our focus is clear: building a scalable, speech-native AI platform that powers global communication, said Mustafa Girgin and M. Sait Demirci, co-founders of Lucida AI. The funding will support Lucida AI's expansion into new languages and markets, further development of its proprietary speech-to-speech AI infrastructure, growth of its enterprise offering and continued product development.

Nomerra, an AI platform for private market operations, has raised $2 million in its first funding round. The round was led by 14Peaks Capital, with participation from Redstone Fintech and senior individuals from firms including KKR and Intapp. The investment comes as private markets are expected to expand significantly over the coming years, while the operational infrastructure supporting the sector has not kept pace. Much of the industry's day-to-day work still relies on emails, PDFs, spreadsheets and disconnected systems, as firms face increasing pressure to manage growing operational complexity with a shrinking pool of qualified professionals. Founded by Johannes Gebendorfer and Jakob Zacherl, Nomerra develops AI software to automate operational workflows across private markets. Unlike public markets, many processes remain fragmented, with limited standardisation and interoperability between systems. The same information is often manually re-entered across multiple applications, sometimes several times for a single transaction. At the same time, operations have become more complex due to new investor channels, more frequent reporting, evolving regulation, semi-liquid fund structures, evergreen vehicles and expansion into new asset classes. Nomerra makes these workflows AI-native, starting with fund accounting, treasury and transfer agency. By integrating with firms' existing systems, its AI agents can access, validate and process data while following each firm's operating procedures. Today, the idea that humans once routed every phone call manually seems absurd. Private market operations are approaching a similar turning point. In a few years, people will look back and wonder how this work was ever done by hand, said Johannes Gebendorfer, co-founder and CEO of Nomerra. Rather than replacing operational teams, Nomerra is designed to shift employees from preparing deliverables to reviewing them. AI agents execute workflows end-to-end before presenting outputs through review interfaces with a complete audit trail showing what was done, why each action was taken and where the underlying data originated. The company will use the funding to expand its engineering team and accelerate product development to meet growing demand from asset servicers and asset managers across Europe and the United States.

Vegvisir, an Estonian defence technology company building the command-and-control software layer for the multi-domain battlefield, today announced a venture investment from Iron Wolf Capital (IWC). Vegvisir's platform addresses one of the most pressing unsolved problems in modern warfare: the absence of a unified, interoperable software layer capable of connecting, visualising, and commanding manned and unmanned systems across ground, air, maritime, and sub-sea domains at operational scale. Unmanned systems have moved from experimental to operational across allied armed forces, but the command-and-control infrastructure required to exploit them effectively remains fragmented, proprietary, and platform-specific. Vegvisir is building the connective layer that bridges that gap: a software-native, platform-agnostic operational interface with AI-driven detection and decision support built in from the ground up — designed to reduce cognitive load on operators managing assets across multiple domains simultaneously.  The company's long-term ambition is to become to allied warfare what air traffic control became to global aviation – the single command interface through which all actors, assets, and decisions flow, regardless of origin or nationality. Proprietary, fragmented command architectures are the single largest obstacle to effective multi-domain operations. Vegvisir intends to make them obsolete. ‍ According to Ingvar Pärnamäe, Co-Founder and CEO of Vegvisir, this investment marks the beginning of the company's next phase, moving from deep product development into operational deployments and commercial scale. "Iron Wolf Capital understands the problem we are solving at a level that goes beyond the financial opportunity. Their footprint at the front line of NATO's Eastern Flank, and their relationships across Baltic and Central European defence and policy circles, give us a strategic amplifier that capital alone cannot buy. This is the partnership we were looking for," The investment draws strong validation from Vegvisir's existing shareholder base, including Kuldar Väärsi, CEO of Milrem Robotics, one of Europe's foremost developers of unmanned ground systems and a key participant in NATO's robotics and autonomous systems programmes, and a personal investor in Vegvisir. "Milrem is building the software-defined robotic systems that future forces will depend on. Vegvisir is building the software layer that makes those systems operationally more capable and easier to adapt. My investment in Vegvisir has always reflected the belief that the future battlefield will be dominated by software-defined systems in which different products and technologies will be interoperable through a shared architecture. The team at Vegvisir has the technology and the ambition to own a software layer which makes the adaptation of robotics seamless at the alliance level," shared Väärsi. ‍ Iron Wolf Capital views Vegvisir as the software layer for future warfare, connecting and multiplying the value of the physical systems that the broader ecosystem is developing. ‍ "We invest where technology meets an irreversible shift in how the world operates. The transition to multi-domain unmanned operations is exactly that kind of shift, and it demands a software-native, platform-agnostic solution that no existing player has adequately built. Vegvisir has the architecture, the team, and the ambition to own that space. We believe that over the next decade, Vegvisir can become one of the defining names in European defence technology, and this investment is our commitment to helping them get there," said Kasparas Jurgelionis, Managing Partner at Iron Wolf Capital. ‍ The investment will accelerate Vegvisir's product development, deepen integrations with allied unmanned platform providers, and expand its pipeline with commercial and government customers across NATO member states. ‍

Italian VC P101 has integrated PranaVentures, an operational venture capital company specialising in Seed investments. The combination brings together more than €600 million in assets raised, creating an Italian firm capable of supporting startups from Pre-Seed and Seed stages through to international scale-up rounds, providing capital, operational skills, an international network, and follow-on capabilities. Today, the combined portfolio includes more than 80 active companies that generated approximately €2 billion in aggregate revenue in 2025 and employed over 5,500 people. Through the integration, P101 expands its investment platform to directly cover the seed segment, further strengthening the team led by Managing Partner Andrea Di Camillo with the highly specialised expertise developed by PranaVentures. The Seed investment strategy will continue to be led by Lisa Di Sevo, Founder of PranaVentures, together with Guido Giordano, Partner, ensuring continuity with the model established since 2021 and with the same distinctive approach to seed investing, based on close collaboration with founders, operational expertise and hands-on support across technology, finance and go-to-market execution. Following the transaction, fundraising has begun for Prana101, the first fund of the new platform, targeting €100 million, with a first closing expected by year-end. Prana101 will focus on Pre-Seed and Seed investments in Italian and European technology startups, backing founders building companies at the forefront of major technology-driven transformations, particularly those enabled by artificial intelligence, next-generation digital infrastructure and innovative services for businesses and consumers. Andrea Di Camillo, Founder and Managing Partner of P101 SGR, commented:  "The integration of PranaVentures strengthens our position in Seed investing and lays the foundations for a new venture capital model in Italy. The market increasingly requires stronger, more specialised and better-capitalised platforms capable of competing at a European level.  Together with the PranaVentures team, we will continue building a more comprehensive, institutional and competitive venture capital platform, with the ambition of surpassing €1 billion in assets under management over time." Lisa Di Sevo, Founder of PranaVentures, said:  "AI has fundamentally reshaped the economics of company building, reducing the capital requirements of early-stage startups by up to 70 per cent while increasing execution speed by as much as eightfold. In this new environment, speed of decision-making and operational efficiency have become as important as access to capital.  Through our partnership with P101, PranaVentures evolves while preserving the principles that have defined our approach since inception — deep seed-stage expertise, hands-on operational support and close proximity to founders. At the same time, we strengthen that model with greater capital resources, enhanced follow-on capabilities and access to a broader institutional and international network.”

Italian VC P101 has integrated PranaVentures, an operational venture capital company specialising in Seed investments. The combination brings together more than €600 million in assets raised, creating an Italian firm capable of supporting startups from Pre-Seed and Seed stages through to international scale-up rounds, providing capital, operational skills, an international network, and follow-on capabilities. Today, the combined portfolio includes more than 80 active companies that generated approximately €2 billion in aggregate revenue in 2025 and employed over 5,500 people. Through the integration, P101 expands its investment platform to directly cover the seed segment, further strengthening the team led by Managing Partner Andrea Di Camillo with the highly specialised expertise developed by PranaVentures. The Seed investment strategy will continue to be led by Lisa Di Sevo, Founder of PranaVentures, together with Guido Giordano, Partner, ensuring continuity with the model established since 2021 and with the same distinctive approach to seed investing, based on close collaboration with founders, operational expertise and hands-on support across technology, finance and go-to-market execution. Following the transaction, fundraising has begun for Prana101, the first fund of the new platform, targeting €100 million, with a first closing expected by year-end. Prana101 will focus on Pre-Seed and Seed investments in Italian and European technology startups, backing founders building companies at the forefront of major technology-driven transformations, particularly those enabled by artificial intelligence, next-generation digital infrastructure and innovative services for businesses and consumers. Andrea Di Camillo, Founder and Managing Partner of P101 SGR, commented:  "The integration of PranaVentures strengthens our position in Seed investing and lays the foundations for a new venture capital model in Italy. The market increasingly requires stronger, more specialised and better-capitalised platforms capable of competing at a European level.  Together with the PranaVentures team, we will continue building a more comprehensive, institutional and competitive venture capital platform, with the ambition of surpassing €1 billion in assets under management over time." Lisa Di Sevo, Founder of PranaVentures, said:  "AI has fundamentally reshaped the economics of company building, reducing the capital requirements of early-stage startups by up to 70 per cent while increasing execution speed by as much as eightfold. In this new environment, speed of decision-making and operational efficiency have become as important as access to capital.  Through our partnership with P101, PranaVentures evolves while preserving the principles that have defined our approach since inception — deep seed-stage expertise, hands-on operational support and close proximity to founders. At the same time, we strengthen that model with greater capital resources, enhanced follow-on capabilities and access to a broader institutional and international network.”

Swiss engineering startup Subatron has obtained €162,000 (CHF 150,000) from Venture Kick to develop a new generation of underwater communication technology.  The solution aims to enable faster, more reliable data transmission beneath the surface, improving real-time connectivity for underwater vehicles, divers, and monitoring systems. Communication underwater remains a major technological challenge. Conventional wireless systems perform poorly below the surface, often offering limited range, slow data transmission, and unreliable connections. These limitations restrict real-time decision-making and reduce operational efficiency in fields such as marine research, infrastructure inspection, offshore operations, and defence. Subatron addresses this challenge with a wireless communication platform designed specifically for underwater environments. Combining proprietary hardware with advanced signal processing, the technology enables stable, real-time data transmission over longer distances than conventional systems. The platform can be integrated into autonomous underwater vehicles. Subatron targets the underwater communication and monitoring technology market with a scalable communication platform for divers, autonomous underwater vehicles, sensor networks, and security applications, and is already working with pilot customers and industry partners across Switzerland and Europe. The funding will support pilot deployments, product industrialisation, certification, international business development, and the expansion of the company’s IP portfolio ahead of its market entry in Switzerland and Europe. “Winning Venture Kick Stage 3 is a strong validation of both our technology and business potential. The program has helped us sharpen our strategy, expand our network, and accelerate the commercialisation of our underwater communication platform,” highlighted CEO and co-founder Samira Baumann. Lead image: Subatron CTO Mathias Werder, CEO Samira Baumann, and CBO Alissa Wyss. Photo: uncredited.

Swiss engineering startup Subatron has obtained €162,000 (CHF 150,000) from Venture Kick to develop a new generation of underwater communication technology.  The solution aims to enable faster, more reliable data transmission beneath the surface, improving real-time connectivity for underwater vehicles, divers, and monitoring systems. Communication underwater remains a major technological challenge. Conventional wireless systems perform poorly below the surface, often offering limited range, slow data transmission, and unreliable connections. These limitations restrict real-time decision-making and reduce operational efficiency in fields such as marine research, infrastructure inspection, offshore operations, and defence. Subatron addresses this challenge with a wireless communication platform designed specifically for underwater environments. Combining proprietary hardware with advanced signal processing, the technology enables stable, real-time data transmission over longer distances than conventional systems. The platform can be integrated into autonomous underwater vehicles. Subatron targets the underwater communication and monitoring technology market with a scalable communication platform for divers, autonomous underwater vehicles, sensor networks, and security applications, and is already working with pilot customers and industry partners across Switzerland and Europe. The funding will support pilot deployments, product industrialisation, certification, international business development, and the expansion of the company’s IP portfolio ahead of its market entry in Switzerland and Europe. “Winning Venture Kick Stage 3 is a strong validation of both our technology and business potential. The program has helped us sharpen our strategy, expand our network, and accelerate the commercialisation of our underwater communication platform,” highlighted CEO and co-founder Samira Baumann. Lead image: Subatron CTO Mathias Werder, CEO Samira Baumann, and CBO Alissa Wyss. Photo: uncredited.

Here in Europe, a new generation of defence startups is emerging far outside the traditional military-industrial complex. Across weekend hackathons, engineers, software developers, drone builders, researchers, soldiers, founders and investors are coming together to prototype technologies for the modern battlefield — many of which are already finding their way to Ukraine. At the centre of that movement is the European Defence Tech Hub (EDTH). I sat down with co-founder Benjamin Wolba to learn how a grassroots initiative became one of Europe's fastest-growing defence innovation communities. From physicist to accidental defence founder Wolba has no background in defence and has never served in the military. He admits his path into defencetech happened almost by accident. He studied physics and completed a PhD in condensed matter physics, researching why materials are magnetic. He admits, “I loved science, but towards the end of my PhD I realised I wasn't going to become a professor, and I didn't want to work for a large corporation. I wanted to build technology with real-world impact, so entrepreneurship felt like the natural next step.” After graduating, he joined Entrepreneur First, although he didn't find the right co-founder at the time, so he later spent two years working as an associate at Lunar Ventures. Following that, he started doing what every good investor tells founders to do — customer discovery. “I explored lots of ideas and spent time figuring out what problem I wanted to solve.” In February 2024, Wolba was in Silicon Valley, networking and trying to understand the startup ecosystem there. At the same time, in El Segundo, near Los Angeles, a defence hackathon was taking place. He admits, “ I was simply fascinated by the idea." He messaged his friend Jonatan Luther-Bergquist, who’s also a physicist and partner at VC firm Inflection. “We looked at what was happening in the US and thought, "Maybe we could organise something like this in Europe. So we organised the first European Defence Tech Hackathon.” ​ Just four months later, that idea became reality. Building a movement from scratch EDTH’s first hackathon took place in Munich in June 2024. Around 150 people attended and built 34 projects covering everything from air defence to demining technologies. Wolba recounts: “We received incredible support. We worked with the Ukrainian Ministry of Defence, Quantum Systems became one of our partners, and many other organisations came on board. Looking back, what surprises me most is that we were complete outsiders. We had no reputation, no previous experience organising anything like this and, honestly, no real reason to expect people would come. Yet the response was overwhelming.” The momentum continued: “After the event, people kept calling me. They'd ask, "Benjamin, do you know an investor?" or "Can you introduce me to someone in defence?" or "Can you help us connect with potential customers?" That was the moment someone said to me, "You're not organising an event anymore—you're building an ecosystem." They were right,” shared Wolba. Today, EDTH has organised 35 hackathons across Copenhagen, Paris, Kyiv, Berlin, London, Prague, Brussels and Tallinn, involving more than 400 teams, with around 40 going on to form companies. inside an EDTH weekend Image: EDTH. Participants register individually — even if they intend to compete as a team — to facilitate security screening, which is a necessity for a defence-focused event. The weekend begins with hands-on technical workshops rather than traditional conference talks, covering topics such as machine learning for defence, data fusion, FPV drone building, battlefield medicine, demining, and electronics. Armed (excuse the pun) with new skills, participants form teams of two to six and spend the next 48 hours designing and building prototypes with few formal interruptions. At the hackathons, there’s huge momentum around drones, autonomy, navigation, target recognition, mesh networking, software-defined radios, and counter-drone technologies, as well as manufacturing automation, logistics, battlefield medicine, and many dual-use technologies. The common theme is solving real operational problems. Organisers provide equipment including 3D printers, soldering stations and electronic components, while many attendees bring their own tools. By Sunday afternoon, teams present working prototypes rather than slide decks, ranging from software applications to drones, sensors, communications systems and other hardware. “That's something I really love about these events. People are incredibly hands-on,” shared Wolba. Investors regularly attend our hackathons as mentors and judges to see promising companies at a very early stage. For founders, it's an opportunity to receive feedback long before they're raising a funding round. The prototypes are then judged by military experts, defence companies and investors. “We always make sure there's Ukrainian representation on the judging panel because operational feedback is incredibly valuable. Afterwards, we send every team detailed written feedback so they can continue improving their projects. For us, Sunday isn't the end of the hackathon. It's the beginning of what comes next.” Image: EDTH. From weekend projects to venture-backed startups While EDTH is not a venture studio, it aims to help founders succeed by connecting teams with investors, testing opportunities, suppliers and defence organisations. “It's intentionally decentralised. Unlike a structured accelerator, founders choose their own path while we help the strongest teams make the right connections.” Unlike traditional events, numerous participants regularly attend several hackathons across Europe and Ukraine as they refine ideas, meet new collaborators and eventually launch companies. Wolba admits this was unexpected. “We thought Munich would be a one-off. Then people asked us to come to Paris, Copenhagen, Warsaw and beyond. The community created the series itself.” Around half of the companies that have emerged have already completed some form of operational testing in Ukraine. Zero Industries is perhaps the best-known success story to emerge from EDTH: “The founders met through our hackathons, worked closely with Ukrainian operators, raised funding and built a real company,” recounts Wolba. The startup develops AI-powered navigation technology that enables autonomous drones to operate reliably in GPS-denied environments. Its proprietary Visual Positioning System (VPS) combines computer vision, advanced mapping, and onboard AI inference to maintain precise positioning even when GPS signals are jammed, spoofed or unavailable. But it is far from the only one, as more companies are being created through the hackathon community, with roughly half of those already completing some form of operational testing in Ukraine. Europe’s capability gap: learning from Ukraine According to Wolba, Europe underestimates the extent to which warfare has changed. “Take Germany as an example. The overwhelming majority of defence spending is still going towards legacy systems. Those systems absolutely have a role, but if you look at where the battlefield is evolving, it's clear we also need to invest much more heavily in entirely new capabilities. We need systems that can defend against drone swarms. We need autonomous systems. We need new approaches to air defence. There are still many things we're getting wrong. It's not just about buying new technology — it's also about understanding how to use it.” Much of that understanding comes directly from Ukraine, where EDTH has now organised three hackathons. The latest became part of Defence Tech Valley and was hosted inside the National Aviation University hangar. Wolba describes Kyiv as the “Champions League” of European defence tech. “Founders build locally, then come to Kyiv to demonstrate their progress and receive direct feedback from operators with frontline experience.” One of the biggest lessons from Ukraine is that tactics evolve incredibly quickly. For example, Ukrainian operators understand exactly how Shahed drones behave. They know the flight profiles, how they approach targets and how to position defensive systems accordingly. “A Shahed drone can fly very low to avoid radar detection or very high to stay out of range of machine guns. Understanding those tactics is just as important as understanding the technology itself. That's why working closely with Ukraine matters so much. They're generating operational knowledge that every European military should be learning from." A movement anyone can join Wolba encourages everyone to get involved: “Come along. Attend a meetup. Join one of our webinars. Come to a hackathon. One of the most important messages we're trying to communicate is that everyone can contribute. You don't have to be an engineer building drones. You might organise events. You might introduce founders to customers or investors. You might write software. You might contribute operational expertise. There are countless ways to help strengthen Europe's defence innovation ecosystem. Our movement has always been built from the bottom up. We haven't waited for governments or relied on government grants. People have simply started building, collaborating and helping one another. That's something everyone can do.” "Europe needs a thousand defence technology startups." In terms of Europe’s capability gaps, Wolba cites air defence as one of the biggest, especially as warfare has shifted from a small number of costly assets to huge numbers of inexpensive autonomous systems. “Europe needs affordable, intelligent systems with better communications, onboard computing and coordination across air, land and sea." Wolba believes that Europe can't afford to wait: “If we're relying entirely on governments or the large defence primes, progress will simply be too slow. What Europe needs is thousands of founders tackling thousands of different problems. We need a thousand defence technology startups. That's how we'll build the capabilities Europe needs for the future.” This week, EDTH is hosting a hackathon in Rome. Next week, EDTH hosts Berlin Defense Tech Week, bringing together founders, engineers, investors, policymakers and military operators for a packed programme of events, including the Berlin Defense Tech Forum, a defence hackathon, meetups, legal and counter-drone forums, and networking sessions across the city.

European semiconductor companies attracted strong investment in 2025 as governments and investors doubled down on technologies underpinning AI, high-performance computing, next-generation communications, and energy-efficient electronics. Funding flowed into companies developing AI chips, photonics, advanced memory, semiconductor materials, chip cooling, and power electronics, reflecting Europe's ambition to strengthen its position across the semiconductor value chain. The Netherlands emerged as the year's leading hub by funding value, driven by NXP Semiconductors' €1 billion European Investment Bank loan and major rounds for Axelera AI, EFFECT Photonics, and Eyeo. The UK also recorded strong activity, particularly in compound semiconductors, photonics, and chip design, while Belgium, Germany, Switzerland, Spain, and France all produced sizeable funding rounds across specialised semiconductor technologies. Debt financing played a significant role alongside venture capital, particularly for established manufacturers and capital-intensive businesses. At the same time, investors continued to back early-stage deeptech startups, with large seed and Series A rounds supporting innovations in optical interconnects, graphene electronics, chip cooling, satellite communications, and semiconductor manufacturing. The year's largest transactions underscored Europe's growing focus on building strategic semiconductor capabilities amid rising demand for AI infrastructure and supply chain resilience. From AI accelerators and memory chips to silicon photonics, advanced cooling systems, and power semiconductors, investors continued to support companies developing the hardware technologies expected to power the next generation of computing and communications (for more detailed analyses of the European technology ecosystem, check out Tech.eu’s annual report: European Tech 2025 - The Big Picture). Here are ten semiconductor companies that raised the most in 2025. Amount raised in 2025: €1B NXP Semiconductors is a Dutch company that develops semiconductor solutions for automotive, industrial, mobile and communications markets. Its portfolio includes microcontrollers, processors, connectivity chips, secure identification technologies, and analogue and power management solutions for applications ranging from connected vehicles to the Internet of Things (IoT). The company works with OEMs and technology providers to support digitalisation, electrification and secure connectivity across multiple industries. In 2025, NXP Semiconductors secured a €1 billion loan from the European Investment Bank (EIB) to support research, development and innovation across its semiconductor portfolio, including power electronics, microprocessors and microcontrollers, at facilities in Austria, France, Germany, the Netherlands and Romania. Amount raised in 2025: €100M Ferroelectric Memory Company (FMC) develops non-volatile ferroelectric memory technologies designed to deliver faster, lower-power and more durable memory solutions. The company commercialises ferroelectric memory IP and semiconductor technologies for AI, edge computing, IoT, automotive and industrial applications. Its technology aims to improve memory performance while reducing energy consumption in next-generation electronic devices. In 2025, Ferroelectric Memory Company (FMC) raised €100 million to accelerate the commercialisation and global rollout of its DRAM+ and 3D-CACHE+ memory chips and system solutions for AI data centres and edge applications. Amount raised in 2025: €61.6M Axelera AI builds hardware and software for AI inference at the edge. Its platform combines purpose-built AI accelerator chips with software tools to enable high-performance, energy-efficient computer vision and generative AI workloads across manufacturing, robotics, retail, healthcare and smart cities. The company focuses on making AI deployment more accessible by delivering high performance with lower power consumption and cost. Axelera AI raised up to €61.6 million in funding in 2025 to develop Titania, a scalable, high-performance, energy-efficient AI inference chiplet for data centres, high-performance computing, and Europe's DARE supercomputing initiative. Amount raised in 2025: $55M Paragraf is a UK company that produces graphene-based semiconductor materials and electronic devices. Its portfolio includes graphene Hall-effect sensors, magnetic sensors and advanced materials for electronics, quantum technologies and industrial applications. The company aims to unlock the commercial potential of graphene in next-generation semiconductor and sensing technologies. Paragraf raised $55 million in a Series C round in 2025 to scale its manufacturing capabilities and increase production capacity, accelerating the mass-market adoption of its graphene-based electronic devices. Amount raised in 2025: $49M Corintis focuses on developing microfluidic cooling technologies for high-performance semiconductor devices. The solutions are designed to integrate directly with chip packages, enhancing thermal management for AI processors, high-performance computing and data centre infrastructure while improving performance and energy efficiency. By tackling one of the industry's key thermal challenges, it aims to enable the next generation of advanced semiconductor systems. In 2025, Corintis raised $49 million across two rounds to scale its microfluidic chip-cooling technology, expand manufacturing, and accelerate deployment for AI data centres and next-generation semiconductor chips. Amount raised in 2025: €33M Swave is a Belgian company that develops holographic display technology based on semiconductor photonics. Its Holographic eXtended Reality (HXR) platform uses diffractive photonics chips to create true holographic images for applications including spatial computing, augmented reality, virtual reality and digital twins. The technology is designed to deliver more natural three-dimensional visual experiences without the need for complex optical systems. In 2025, Swave Photonics raised €33 million across two rounds to commercialise its holographic display technology for augmented reality smart glasses and heads-up displays. Amount raised in 2025: £25M CamGaN Devices designs gallium nitride (GaN) power semiconductors for energy-efficient power conversion. Its technology is intended to improve efficiency, reduce power losses and shrink the size of power electronics used in data centres, electric vehicles, consumer electronics and industrial systems. The company's solutions are designed as drop-in replacements for conventional silicon-based power devices. In 2025, GaN Devices raised £25 million in a Series C round to scale its gallium nitride (GaN) semiconductor technology, expand operations internationally, and accelerate the deployment of energy-efficient power devices for industrial, data centre, and automotive applications. Amount raised in 2025: $30M Salience Labs is a UK company that develops photonic computing technology for AI infrastructure. Salience Labs designs optical interconnects and photonic systems that improve data movement and energy efficiency in AI accelerators and high-performance computing, helping address bandwidth and power constraints in next-generation data centres. The company focuses on enabling faster and more scalable AI computing through integrated photonics. In 2025, Salience Labs raised $30 million in a Series A round to accelerate the development and commercialisation of its silicon photonic optical switches for large-scale AI data centre connectivity, enabling higher bandwidth, lower latency, and reduced power consumption. Amount raised in 2025: €25M CamGraPhIC develops graphene-based silicon photonics technologies for high-speed optical communications. These solutions combine graphene with silicon photonic devices to improve data transmission, reduce energy consumption and enable faster optical interconnects for AI, data centres and telecommunications. The company aims to overcome performance limitations of conventional silicon photonics using graphene-based components. CamGraPhIC raised €25 million in a Series A round in 2025 to scale its graphene photonics technology for AI and data communications. Amount raised in 2025: £18M IQE is a supplier of advanced semiconductor epitaxial wafers used in compound semiconductor devices. Its materials support applications across wireless communications, photonics, power electronics, quantum technologies and advanced sensing, supplying semiconductor manufacturers worldwide. The company specialises in engineered wafer technologies that enable high-performance electronic and optoelectronic devices. In 2025, IQE raised £18 million through convertible loan notes to provide short-term liquidity and working capital while completing its strategic review.

The AI industry has spent the past several years learning a critical lesson: better data often matters as much as better models. While advances in large language models have been powered by increasingly sophisticated datasets and data pipelines, robotics has yet to undergo the same transformation.  Robotics teams are working with vast quantities of video, sensor data, and demonstrations, but much of the infrastructure needed to process, annotate, and improve that data remains immature. Macrodata Labs believes that closing that gap could become one of the most important challenges in robotics AI. Macrodata Labs recently emerged from stealth, launching Refiner, an open-source framework and cloud platform for processing robotics datasets.  The company raised $4 million in pre-seed funding in June this year to build infrastructure for the robotics data loop. The round was led by Air Street Capital, with participation from Drysdale Ventures, OPRTRS club, Kima Ventures, YG (Alex Yazdi), >commit, Thomas Wolf, and business angels from some of the world’s leading AI labs and technology companies.  Macrodata Labs builds infrastructure for the robotics data loop. Its first product, Refiner, is an open-source framework and cloud platform for robotics data processing, helping teams turn raw physical-world data into better training datasets.  I spoke to the CEO and co-founder, Guilherme Penedo, to find out more. From  building LLM datasets to building robotics infrastructure Macrodata Labs was founded by Guilherme Penedo and Hynek Kydlíček, who formed the core team behind several of Hugging Face's largest open LLM dataset efforts. They created widely used datasets such as FineWeb, FineWeb2, FinePDFs, and FineTranslations, which have been used by teams at NVIDIA, Google, AI2, and Z.ai, and contributed to large-scale training projects such as Open-R1 and SmolLM. Penedo was part of the team behind Falcon, one of the strongest open-source models at the time of its release. After that, he joined Hugging Face, where he focused on building large-scale datasets for training AI models. “That's where I met my co-founder, Hynek Kydlíček. We worked together on projects such as FineWeb, which processes large portions of the internet and turns the data into high-quality training datasets. FineWeb became one of the most widely used open datasets for language model training, and we later expanded that work into other areas, including PDFs and multilingual datasets.” The common theme throughout their work was figuring out how to take massive amounts of raw data and transform it into something that can produce significantly better AI models. While building large-scale datasets at Hugging Face, the founders saw that progress was not only about model architectures or compute, but also about the infrastructure needed to collect, transform, inspect, and iterate on training data at scale.  After seeing how better data infrastructure helped unlock progress in LLMs, the founders believe robotics is approaching a similar inflection point.  Why better data, not better models, could unlock robotics While advances in LLMs and vision-language models (VLMs) are making robots increasingly capable, the data layer underpinning robotics remains underdeveloped. Physical-world data is larger, messier, more fragmented, and far more difficult to transform into useful training datasets than text. Penedo explained: “In language models, we learned how difficult it is to transform raw data into datasets that consistently produce high-quality results. We realised that robotics is facing many of the same challenges, but on an even larger scale.” According to Penedo, the key difference is that many data-processing tasks in language models can be handled with relatively simple rules, whereas robotics requires far more interpretation. “You might have hundreds of hours of video showing humans performing tasks, but before that data becomes useful for training robots, you need to understand what is happening in the scene,” he said. “For example, if someone is washing dishes, you need to identify individual subtasks: picking up a plate, applying soap, rinsing, and so on. You may also need to estimate hand positions, infer actions, and map human movements to robotic equivalents.” The challenge extends beyond understanding actions. Robotics datasets combine video, sensor streams, trajectories, and other multimodal inputs, creating large, complex datasets that are difficult to store, process, and standardise. Different robotics companies often use their own data formats and workflows, while many questions about what data should be collected and how it should be annotated remain unresolved. “We believe robotics is the next major frontier for AI,” said Penedo. “The progress we've seen in large language models and vision-language models is now enabling a new generation of robotic systems. At the same time, robotics is increasingly benefiting from the same scaling principles that transformed language models: better data leads to better models.” As a result, a significant amount of work is required to label, annotate, filter, and enrich data before it becomes useful for training. “These constraints make data work in robotics especially important,” Penedo said. “Teams need scalable, reliable tooling so they can process demonstrations, test new annotations, and iterate on datasets without rebuilding their data stack every time they change embodiment, sensors, data format, or labeling method.” Penedo cautions that the industry is still very early, with many companies investing heavily in collecting more data, improving model architectures, or building better hardware.  “Those things are important, but comparatively little attention has been paid to improving the quality of existing data. Many teams still rely on manual processes for annotation and data preparation, even though modern AI systems can automate much of that work. The data you collect today will likely remain valuable across multiple generations of models and architectures. That's why we think infrastructure for data processing is one of the most important pieces of the stack.” Refiner: infrastructure for the Robotics data loop Robotics companies are often hardware-first organisations, but Macrodata Labs believes that the software laye r— and specifically the data layer — is what will ultimately determine how capable these systems become. Refiner offers an open-source framework for processing robotics datasets. It enables robotics teams to ingest data, process demonstrations, and run workflows such as hand-tracking, subtask annotation, and reward model scoring. The framework supports a wide range of robotics data formats and can process multimodal robot episodes — including trajectories, camera streams, sensor data, and annotations — within a single pipeline. Designed to work directly with cloud storage, it allows teams to work with large datasets without first downloading them locally. Penedo explained: “Users don't need to download terabytes of data locally before they can start working. Refiner can stream data directly from cloud storage, process it efficiently, and run workflows across distributed infrastructure.” Refiner also supports GPU-based processing, which is increasingly important as robotics data pipelines rely on AI models for tasks such as annotation, understanding, and evaluation. The broader goal is to make robotics data infrastructure more accessible and scalable while giving teams the flexibility to work across different robots, sensors, and workflows. Through the hosted Macrodata Labs platform, users can scale the same pipeline from local Python execution to managed cloud compute without rewriting their workflows. The platform handles orchestration, scheduling, CPU and GPU workers, data traceability, failure recovery, and observability, while customers pay only for the compute resources they use. Right now, the company is focused on robotics companies that train models and build robotic systems. Over time, Penedo predicts the market will expand: “As robotics models become more capable and accessible, we expect more organisations to buy robots off the shelf and fine-tune them for specific tasks. At that point, we can help those customers understand what data they need to collect and how to adapt models to their environments. But today our primary customers are the teams building the underlying robotics systems.” Building a robotics startup in stealth I was curious what it was like for the team building a company in stealth.  Penedo admits that there were definitely challenges. “When you're operating in stealth, people can't easily look you up online or validate what you're doing.  That means introductions and personal networks become much more important because potential customers and partners don't have much public information to work with. That said, we never intended to remain in stealth for long.  The goal was simply to give ourselves a few months to build the first version of the product, validate the core ideas, and begin working with early users before going public.” Why Europe can lead the next wave of robotics Macrodata is technically structured as a US company, largely for fundraising reasons, but based in France and would love to see Europe become a major force in robotics. Europe is frequently cast as trailing the US in AI, but Penedo believes robotics is one area where Europe remains highly competitive. “You see strong clusters around Zurich, driven by ETH Zürich and the companies emerging from that ecosystem. Munich is another major centre. More broadly, Europe remains highly industrialised and has a large manufacturing base, which creates real demand for innovation in robotics. That gives Europe an opportunity to play a significant role in this next wave of AI.” Macrodata Labs' immediate focus is helping users adopt Refiner and gathering feedback from the robotics community, while investing heavily in research into how better data pipelines can improve model performance. “We want to go beyond making robotics data processing more efficient and explore how better data pipelines can actually improve model performance. That means testing new approaches, training models, running experiments on real robotic systems, and continually measuring whether our methods produce better outcomes,” shared Penedo. 

Last week, we tracked more than 75 tech funding deals worth over €2.1 million and over 5 exits, M&A transactions, rumours, and related news stories across Europe. ? The top three industries that raised the most were fintech (€502.9 million), security (€500 million), and semiconductors (€349.7 million). At the country level, ?? Germany took first place (€806.8 million), followed by ?? France (€625.2 million) and ?? the Netherlands (€372.7 million). Last week, we announced the launch of the Tech.eu Funding Explorer, now free and open to everyone. Built for founders, investors, analysts, journalists and others following the European tech ecosystem, the platform provides access to funding data, investor activity, company profiles and market trends. As the Funding Explorer is currently in beta, we welcome your feedback through the dedicated feedback section within the platform. Your input will help us refine and improve the product as we continue to develop it. ❗ Now, let's get you up to speed on everything that happened last week, including your handy.csv file, allowing for an even more in-depth analysis. Have a great week! Funding deals by amount GERMANY: Stark bags €500M in new funding FRANCE: Health insurance outfit Alan agrees €480M funding round NETHERLANDS: Chip gear maker Nearfield Instruments raises $380M GERMANY: Car subscription company Finn becomes a unicorn after a €140M Series D funding round GERMANY: Taktile receives $110 million in a Series C FRANCE: Tissium raises €30M in equity funding, €30M in debt NETHERLANDS: Leyden Labs secures €40M to develop intranasal protection against influenza and coronaviruses UK: Isometric raises over £30M to expand AI certification platform FRANCE: Tsuga lands €30M to expand AI agent platform UK: Astral Systems raises £23M to tackle the global shortages of medical radioisotopes SPAIN: a16z backs Prosper AI with $30M as healthcare providers seek fewer admin tools UK: Tech firm Seat Unique secures £20M investment SWEDEN: Sportway raises €20M as AI reshapes sports broadcasting UK: equipal raises £16.25M in funding UK: Superlight raises £15.9M to boost commercial EV truck manufacturing GERMANY: VARM bags €17.5M to scale insulation across Europe UK: HICX secures £15M to close supplier data gap FRANCE: AlpSemi raises €17M to advance solid-state circuit breaker technology GERMANY: Almetra secures €16.3M Series A to drive smarter manufacturing GERMANY: JUPUS raises €13M to power the next generation of AI-driven law firms FRANCE: Flease secures €13M to expand sustainable fleet leasing across France UK: New funding takes Trimtech Therapeutics’ seed round to £35.6M BELGIUM: Timefold raises $13M Series A to scale scheduling optimisation infrastructure IRELAND: Ubotica raises $11M to scale real-time maritime intelligence from space SPAIN: FOSSA raises €9.25M to accelerate the deployment of secure communications satellites FRANCE: Linc raises €8.5M with 100+ angels to crack world's most complex payroll market FRANCE: Wheere raises €8.5M to launch its first satellite IRELAND: TensorX raises €8M in seed funding FRANCE: Sopht raises €7.5M to reduce companies' IT bills SPAIN: H2SITE raises €6M in second close of Series B funding UK: Silveray secures £5M to expand X-ray technology into healthcare NORWAY: Eqon raises $6M in seed funding AUSTRIA: Graph Therapeutics brings total funding to over $10M to advance precision immunology CZECH REPUBLIC: Logistics startup Grid.online lands €4M after growing deliveries 10× in a year GERMANY: Maple Aviation secures a €4M investment AUSTRIA: Fintech Talentir secures €4M for international expansion GERMANY: Mona AI secures a €3.8M investment as part of a seed extension SWEDEN: Fika Jobs raises $4M for AI-powered video resumes AUSTRIA: Ora Computing raises €3.5M to build the efficiency layer of the AI stack GERMANY: Companisto, Prolimity Capital Partners, and others are investing €3.3M in Innok Robotics ITALY: Compri secures €3.2M to build AI-powered procurement teams GERMANY: Kyrok secures €3.1M to bring AI to pharma and chemical supply chains GERMANY: mkind receives a €3M investment GERMANY: Zelara lands €3M to bring continuous learning to customer engagement SPAIN: Kalipso raises $3.2M to scale regulatory compliance platform DENMARK: Acodyne secures €2.5M to develop next-generation autonomous logistics aircraft SWEDEN: Wayout raises €2.42M to scale decentralised drinking water infrastructure platform REPUBLIC OF SERBIA: SuperPlane secures $2.6M to turn production operations into an AI-native workflow layer GERMANY: Wakeline lands €2.1M to bring continuous learning to AI SPAIN: Floox closes a €2M funding round to consolidate its position as a European leader in the electric mobility sector UK: Ademen has secured £1.6M investment UK: Adtech Covatic raises £1.5M IRELAND: HR Duo secures £1.4M to scale and accelerate UK expansion DENMARK: Serpier raises €1.4M to help online retailers improve digital visibility TÜRKİYE: Ideasets received a $1.25M investment from Gelecek Holding, based on a valuation of $10M UK: AI company Unloqs secures £900,000 funding SPAIN: WAIIS raises €1M to boost its shared mobility super app GERMANY: CoTrainer closed €1M investment round SPAIN: Naturr closes an investment round of €620,000 LITHUANIA: Superpal raises €500,000 for AI coworker platform built inside Slack UK: LabCycle secures £430,000 to commercialise lab plastic recycling system and cut incineration waste SWEDEN: MiMove secures €450,000 to expand real estate platform across Southern Europe SWEDEN: Waveium raises €470,000 pre-seed round SWITZERLAND: Isospec Analytics was awarded a €434,000 Tech Growth loan from FIT SWITZERLAND: Backbone received a €433,000 FIF Growth Award TÜRKİYE: Revogo, a sustainable e-commerce platform, has received a $300,000 investment SPAIN: SegurosIA raises €150,000 from Eoniq.fund to boost its artificial intelligence technology for insurance companies SWITZERLAND: Crìa Technologies received a €108,000 FIF Seed Tech Award SWITZERLAND: Skilder received a €22,000 Digital Grant by FIT SWITZERLAND: MYSTONES received a €22,000 Digital Grant by FIT PORTUGAL: Coalex.ai secures funding to scale ‘decision firewall’ for AI in production UK: URUNN app secures seven-figure raise AUSTRIA: TradersYard raises fresh capital SWEDEN: Anferra receives investment TÜRKİYE: Cypien AI, having completed its pre-seed follow-up investment, has reached a valuation of $4.5M POLAND: Talkin’ Things lands Orbit Capital financing amid growing traceability demand GERMANY: Backed by Lakestar, Seedcamp and EWOR, SE3 unveils spatial AI platform for autonomous systems GERMANY: lingomatch received an undisclosed sum of investment Exits and M&A activity POLAND: LiveKid acquires Aldea to expand in Latin America GERMANY: House of Gaia Group is acquiring Codio Impact GERMANY: The young Munich-based legal tech company beglaubigt.de is acquiring the insolvent company firma.de FRANCE: Withings acquires the Rennes-based medtech company Biosency UK: Edtech Pastest acquires financial platform Medics’ Money UK: PayPoint acquires Aperidata NETHERLANDS: Backbase buys agentic banking platform Kasisto

This week, we tracked more than 75 tech funding deals worth over €2.1 billion and over 5 exits, M&A transactions, rumours, and related news stories across Europe. If email is more your thing, you can always subscribe to our newsletter and receive a more robust version of this round-up delivered to your inbox. Either way, let's get you up to speed. ? Notable and big funding rounds ?? Stark bags €500M in new funding ??  Health insurance outfit Alan agrees €480M funding round ?? Chip gear maker Nearfield Instruments raises $380M ??‍?? Noteworthy acquisitions and mergers ??  LiveKid acquires Aldea to expand in Latin America ?? House of Gaia Group is acquiring Codio Impact ??  Withings acquires the Rennes-based medtech company Biosency ? Interesting moves from investors ? Revolut and Synthesia early backer Seedcamp raises $320M, invests in US ?  Main Capital doubles down on enterprise software with €5.25 billion fund close ? New €34M Nucleo Ventures fund targets startups and SMEs across Central and Eastern Europe ? Blue Lake VC closes in on first fund with British Business Bank backing for immigrant-led startups ? ARX Robotics and Roboneers form ARX Industries to scale unmanned ground vehicle production ? European tech startups to watch  ??  SuperPlane secures $2.6M to turn production operations into an AI-native workflow layer ?? Wakeline lands €2.1M to bring continuous learning to AI ?? Serpier raises €1.4M to help online retailers improve digital visibility ?? Superpal raises €500,000 for AI coworker platform built inside Slack ?? LabCycle secures £430,000 to commercialise lab plastic recycling system and cut incineration waste

As an enthusiastic proponent of AI, one might expect the UK’s AI minister to be a keen user of ChatGPT or other popular AI chatbots in the course of his ministerial duties- perhaps to make his days more productive or to stay abreast of the latest technological advancements. But Kanishka Narayan has not used ChatGPT, which has over 1bn active users, Claude, or any other popular AI chatbots for ministerial business since he became AI minister in September last year, according to a Freedom of Information (FoI) request. The request also reveals that Technology Secretary Liz Kendall has not used ChatGPT or other AI chatbots for ministerial business since being appointed to the role in September 2025. The FoI was requested on April 23. Narayan, a former VC who has worked in Silicon Valley, however, does use ChatGPT for “core research” and background information for personal use, the Telegraph has previously reported. Narayan has urged MPs to stop writing their parliamentary speeches with ChatGPT, following analysis showing that MPs were using the OpenAI-made chatbot to write speeches increasingly frequently. Like Narayan, Kendall also uses AI outside work. “Well, I use AI personally rather than at work, I’ve got to be honest,” Kendall previously told the BBC. Peter Kyle, Kendall’s predecessor, used ChatGPT for policy advice and ask what podcasts he should appear on. While Narayan and Kendall might be shunning AI chatbots for ministerial business, government officials are expected to use AI. The government is also deploying AI to modernise public services and increase civil service productivity. The government says its AI tool Consult, designed to speed up public consultation, has sped up analysis of over 50,000 responses to a government-commissioned review of the water sector. A government spokesperson said: “The Secretary of State and AI minister do not personally use chatbots in the course of their ministerial business, but we are using a range of AI tools to support the work that they direct. AI has the potential to save time on routine tasks and cut through admin for civil servants. Ministers are focused on steering that work, and making the final decisions- as the public would expect."

The CEO of UK-based spacetech investor Seraphim Space says Europe is “catching up” to US spacetech, powered by the SpaceX IPO, increased deal sizes, and European spacetech startups pivoting to defence applications, helping drive up investment.   Speaking on the Tech.eu podcast, Mark Boggett, co-founder and CEO of Seraphim Space, discusses the recent SpaceX IPO, its impact on the European spacetech sector, as well as the investment landscape, and European sovereignty.   Boggett also talks about Seraphim's portfolio, which includes its biggest holding in aerospace company Iceye, which is likely eyeing an IPO, and Hawkeye 360, the signal intelligence company.   Boggett said around 80 per cent of Seraphim's portfolio revenues were now coming from defence, as spacetech startups pivot to defence. While still lagging behind the US in terms of the size of the investment deals, Europe is catching up, he said, as witnessed by some big-ticket spacetech deals this year.    Iceye raised over €1bn in a funding round valuing the company at around €10bn, while German rocket developer Isar Aerospace announced a €270m funding round. On the impact of SpaceX's IPO on the broader sector, Boggett said it will bring a wider range of investors to the sector, such as big institutional investors.   He also said a key driver of optimism in the sector was Starship, SpaceX’s new launch vehicle, which he said was a “game-changer”, making it easier to launch infrastructure into space.   He said: “The price and scale of launch are now changing. It is now opening up the market for mega infrastructure in space because previously it has been cost-prohibitive to put large infrastructure into the space environment.” Image: NASA on Unsplash

Louise Ballard, co-founder and CEO of Atheni AI, has a mission — to ensure that when it comes to AI, nobody gets left behind. “As AI becomes embedded across every profession, we don't want to create a two-tier society where only those who can afford expensive tools or specialist training are able to benefit. Everyone should have access to the knowledge and confidence they need to use AI well.” When it comes to AI startups, Atheni AI is an outlier. Its female founders, Louise Ballard and Mackenzie Howe, aren’t postdoc academics or 20-something men vibecoding their way to customers.  Ballard spent three decades in corporate communications before selling her PR agency to Huntsworth in 2009. After recovering from two cancers, she reconnected with future co-founder Mackenzie Howe, an entrepreneur and former institutional investment consultant. Together they realised the biggest challenge around AI wasn't the technology—it was helping people use it effectively. And what they’ve created is Atheni.ai, a UK startup focused on successful AI adoption rather than AI model development. It helps organisations ensure that employees actually integrate AI tools such as ChatGPT, Claude, and Microsoft Copilot into their daily work in ways that improve productivity and decision-making.  The AI adoption problem Ballard contends that “Anyone can open ChatGPT and ask it a question. The real question is whether they know if the answer is good. Can they provide the right context? Can they challenge the output? Can they connect different tools together? Can they build simple automations? Those are skills anyone can learn, but only if they have the confidence and support to explore them." Talking to former clients across a range of industries,  everyone described the same challenge. Ballard explained that they'd bought AI licences, rolled them out across the company and expected people to use them, but adoption was low. “Those who did use the tools often weren't using them effectively, while many simply reverted to the way they'd always worked." That experience reflects a broader industry trend. A global study released this week by CambrianEdge.ai, which surveyed 775 AI users across 104 organisations, found that 55 per cent of professionals see isolated individual use of AI and the lack of structured human-AI workflows as the biggest barrier to adoption. The study found that more than a quarter of organisations still lack basic collaboration infrastructure, such as shared prompt libraries, training, and quality standards, while 18 per cent have already scaled back AI initiatives due to poor adoption and inconsistent results. The findings mirror research from BCG, which found that although 96 per cent of 300 global CMOs say AI is driving business transformation, almost half still use it only for isolated tasks rather than embedding it across workflows. Organisations with comprehensive AI infrastructure — including shared tools, training, prompt libraries and governance — were dramatically more likely to report significant business impact. “We quickly realised that real transformation doesn't happen because you have one or two AI champions. You need the whole team developing confidence together,” shared Ballard. Why training isn't enough Ballard argues that experienced workers often become the strongest AI users because they bring judgment that AI cannot replace. Some of the biggest barriers to adoption are fear of being replaced, habit — “If you've completed a task the same way for 10, 15 or 20 years, changing that workflow takes cognitive effort—especially when you're already busy.” and, importantly, relevance. Ballard contends that traditional training, like workshops and demos, is too generic because you attend or watch something and think, "That's interesting." Then you return to work and immediately go back to your existing processes. AI is different because it's deeply personal. Even two people doing almost identical jobs will use it differently. "We saw this repeatedly. We'd run workshops, everyone would leave enthusiastic, and six weeks later, clients would tell us very little had actually changed. That's when we realised this wasn't primarily a training problem — it was a coaching problem. People need ongoing guidance while they're working.” Coaching people while they work Atheni sits alongside people while they're working, understands their role and guides them through what the company calls the Atheni capability scale — from Curious through to Pathfinder. Rather than delivering generic lessons, it coaches individuals based on how they're actually using AI, the quality of their prompts, the workflows they're building and the opportunities they're missing. Through a browser-based assistant and analytics dashboard, the platform provides personalised coaching, tailored learning missions and practical recommendations based on an individual's role, helping teams progress from basic AI adoption to more advanced, strategic use.  It also gives organisations visibility into team-wide AI capability and adoption, enabling them to build AI literacy and drive meaningful behavioural change rather than simply measuring tool usage. Success for clients of Atheni is ultimately adoption. For example, one client realised some people were using free AI tools independently and potentially exposing confidential information, so they wanted a proper strategy, while others described themselves as technophobes. “Over three months, we worked with them consistently. By the end, they'd reached around 90 per cent adoption, with roughly a third of employees progressing into our highest capability tier. What changed wasn't simply that they were using AI more often. They understood what good usage looked like. They realised AI wasn't replacing their expertise—it was extending it. They could stress-test ideas, explore scenarios, analyse information and solve problems they simply couldn't have tackled manually.” Access doesn't equal adoption Atheni works with organisations before, during and after AI rollouts, helping them move beyond simply distributing licences. “The biggest misconception is thinking that access equals adoption. Giving someone a Copilot or ChatGPT licence doesn't automatically change the way they work.” Ballard contends that it's critical to think about how the work gets done in the first place. For example, a client may want to use AI to free their people to spend more time thinking, solving problems, and working with customers, rather than getting buried in repetitive processes. “One corporate finance client redesigned a monthly spreadsheet process rather than simply automating it. That's the real shift. It's not just efficiency — it's redesigning work.” She contends that the companies that struggle are often those that have simply rolled AI out across the organisation and assumed adoption would happen naturally. “They find themselves asking why everyone has a licence, but nothing has really changed, or why they're suddenly generating lots of AI-written emails that don't actually communicate anything particularly well.” AI is like "driving a Ferrari to the supermarket" Ballard likes to compare AI to "driving a Ferrari to the supermarket". People own incredibly powerful technology but use only a tiny fraction of its capabilities. "Success isn't measured by how many prompts someone writes every day. It's measured by how deeply AI becomes integrated into the way they think and work. Someone who uses AI only twice a day but has built sophisticated workflows creates far more value than someone who spends the day asking it to rewrite emails." That's the difference between depth and volume, and that's what Atheni is designed to develop. Building AI around how people actually work Ballard believes that company expertise is essential to successful AI adoption: “I've spoken to organisations that have brought in sophisticated AI systems designed entirely by external technology teams, only to find they don't reflect how the business actually operates. They end up rebuilding everything because the people designing the workflows didn't understand the day-to-day reality. Increasingly, every professional will need two complementary skill sets. One is their domain expertise—whether that's journalism, finance, marketing or law. The other is enough AI literacy to build and adapt the tools they need themselves. Rather than relying on a central technology team to solve every problem, people should be empowered to create solutions that fit their own workflows.“ Preparing for an agentic future Of course, the risk is that AI systems become so intuitive that workers no longer need a platform like Atheni. Ballard admits it's a question the company considers often and asserts that the skills people need will continue to evolve alongside AI. “Today, we spend a lot of time helping people write better prompts and understand how to work effectively with AI. In a few years, prompting may no longer be the key skill. Instead, people may be building increasingly sophisticated AI agents or orchestrating multiple systems together. The underlying challenge doesn't disappear — it simply changes. You could use Atheni Ai in the future to coach negotiation skills, leadership, intergenerational communication or any workplace capability where people benefit from ongoing, contextual guidance while they're actually doing the work.” Closing the credibility gap Atheni.ai raised £350,000 in May this year. Ballard admits fundraising was one of the hardest parts of building the business. She admits that when they first started raising, the founders assumed investors would immediately understand the problem they were solving. "Instead, many questioned whether the problem even existed. The reaction was often, 'AI is easy to use. Why would anyone need coaching?'" The team paused fundraising, raised a small friends-and-family round and focused on building the product. Ballard admits: "As a female founder in my fifties, I realised we weren't just facing a funding gap—we were facing a credibility gap. We're two women who don't fit the stereotype of AI founders, despite years of experience building businesses and working with technology. That meant people often underestimated both the problem and our ability to solve it. "I'd built and sold a business and advised CEOs throughout my career, yet suddenly we were having to prove our credibility in ways I'd never experienced before." Finding the right investors changed everything. Once they connected with people who understood the opportunity, Atheni closed its round in around six weeks. "Because we're already generating revenue through our consulting work, we've been able to validate the problem before scaling the software platform. Traditionally, there was a clear distinction between consulting and software businesses. Today, particularly in AI, that distinction is becoming much less relevant. If you're solving complex human problems, you need deep domain expertise, and that expertise often comes from working directly with customers before it's embedded into software. That's exactly what we've done." Ballard hopes that, as AI reshapes the workplace, humans remain at the centre. "We have an opportunity to shape what the future of work looks like, and I'd like that future to be one where humans remain firmly at the centre." Lead image: Louise Ballard and Mackenzie Howe.

Spatial AI company SE3 Labs today emerges from stealth. SE3 builds foundational spatial intelligence technology for the next generation of autonomous systems. As autonomous systems operate in increasingly complex and contested environments, the ability to perceive, reason, and act in three dimensions becomes the defining capability.  SE3 provides the spatial intelligence layer that sits between raw sensor data and meaningful action.  The company is backed by Lakestar and Seedcamp, alongside EWOR, the Sequoia Scout Fund, UnternehmerTUM Funding for Innovators, SDAC, Magnetic, TwinTrack Ventures, Plug and Play, the founders of Flixbus and Ascending Technologies, and strategic angel investors.  Teaching autonomous systems to understand the physical world Today's unmanned aerial vehicles can fly a pre-programmed route or follow a beacon, but they cannot reason about the world around them. SE3 closes that gap.  Its stack operates as a single platform and domain-agnostic system across aerial, ground, and mixed swarms, and is already under contract with the German Bundeswehr.  The stack is hardware-agnostic, modular, and runs on-edge. Customers can integrate SE3 as a standalone capability into an existing platform or have SE3 cover the full software stack for UAVs    "Advances in AI have enabled machines to understand language. The next step is enabling them to understand the physical world," said Lukas Köstler, CEO and co-founder of SE3.  "We’re building the category leader in Spatial AI. Our technology unlocks a new generation of autonomous systems that can operate reliably in real-world conditions across defence, public safety, and industrial applications." According to Dr Klaus Hommels, Founder and Chairman of Lakestar,  Europe’s sovereignty depends on its ability to build and scale the critical capabilities that will define the next era of defence.  “SE3 is developing one of those capabilities, and we believe it will be essential to Europe’s security, resilience and technological independence.” Spatial autonomy turns raw sensor data into shared understanding  SE3's navigation system delivers precise, continuous autonomous navigation in GPS-denied terrain and under active electronic warfare.  Onboard visual-inertial odometry and real-time map matching replace the satellite link: each platform builds and updates its own spatial picture as it moves, and paths adjust dynamically as the environment evolves, without operator intervention.  On top of the resilient navigation foundation, the perception stack turns raw visual data into a continuously evolving 3D understanding of the operating environment. Terrain, elevation, and points of interest are reasoned across in real time, and objects are localised in three-dimensional space to sub-metre accuracy. That spatial picture is shared across the swarm, so every platform converges on the same target. For the operator, it forms a live 3D common operating picture of the environment. The operator interacts with that picture in natural language: spoken intent reaches directly into the 3D space and is translated into action across the swarm. AI agents run continuously on top of the same picture, analysing the real world in real time and conducting advanced reasoning over what they see.   From that shared 3D picture, one operator commands the swarm. This turns a single operator into a force multiplier: mixed aerial and ground platforms are directed in natural language, and operator intent is interpreted spatially and distributed as behaviour across the swarm, so the size of the force scales with the mission rather than with the headcount available to fly it.  “Perception alone doesn't make a system autonomous. The harder problem is discernment: knowing what matters, separating the urgent from the merely important, and ranking intentions against the mission and the environment. Without that layer, more sensors just produce more noise. SE3 is building the discernment layer for physical AI.” said Carlos Eduardo Espinal, Managing Partner at Seedcamp. Built on world-class research from the leading institutions in computer vision and AI  SE3 was founded by Lukas Koestler, Simon Klenk, and Daniel Cremers, bringing together expertise in robotics, computer vision, artificial intelligence, and defence technology. The founding team has produced more than 400 scientific publications with over 87,000 citations, combining frontier AI research with experience from organisations including Nvidia, Tesla Autopilot, Boston Consulting Group, Skydio, and Isar Aerospace. Cremers, who holds the Chair for Computer Vision and Artificial Intelligence at the Technical University of Munich, is president of the European Computer Vision Association, Director of the Munich Centre for Machine Learning, a member of the Bavarian Academy of Sciences and Humanities, and a recipient of the Gottfried Wilhelm Leibniz Prize.  The company is committed to putting cutting-edge spatial AI in the hands of the German Bundeswehr and allied forces operating in modern conditions. SE3 participates continuously in military exercises across Europe, and in recent exercises has reduced the sensor-to-shooter timeline by an order of magnitude. The technology is also being validated in real operational settings today. The company has won multiple defence contracts and is operational with the Bundeswehr.  “Many teams excel either at frontier research or at operational deployment. SE3 is unusual because it brings both together in one culture. The team is deeply technical, close to the customer and unusually low-ego, yet the ambition is enormous: to build technology that becomes a core part of how autonomous systems operate in the physical world,” said Enrico Mellis, who co-led Lakestar’s investment in SE3. 

German company ARX Robotics and Ukraine's Roboneers today announced the formation of ARX Industries, a joint venture to industrialise the mass production of unmanned ground vehicles. The new entity will produce the Rys Pro UGV at scale, delivering software-defined ground capabilities to the Ukrainian Defence Forces.   The joint venture brings together ARX Robotics and Roboneers' proven expertise in unmanned ground vehicle production, industrial scale, and operational deployment, forming a partnership designed to meet the urgent and growing demand for UGVs on the modern battlefield. The partnership directly supports Ukraine’s target of supplying 50,000 UGVs to the military in 2026.   ARX Industries will produce the Rys Pro UGV across facilities in Germany and Ukraine, combining industrial capacity with frontline proximity. This industrial complex will ensure both the scale and the agility to respond to evolving operational needs.   The joint venture aims to produce thousands of Rys Pro units within the first year of operation. Output is projected to scale progressively, reaching tens of thousands of units annually in subsequent years. The UGVs will be deployed across mission-critical roles covering casualty evacuation, frontline logistics - transporting supplies, medical aid, mission-specific modules including mining and demining - and combat configurations.   As part of this joint venture, ARX Industries will cover manufacturing, maintenance, and operational support, creating a long-term production base for sovereign UGV capacity. This will strengthen European partners' ability to design, manufacture, sustain, and deliver critical unmanned ground systems at scale.   The cooperation is supported by both the Ukrainian and German governments, underscoring the strategic importance of the joint venture for European defence sovereignty. Established under the Build with Ukraine initiative, the partnership accelerates the delivery of battle-proven UGVs directly to the Ukrainian Defence Forces, strengthening Europe's defence posture.   At ARX Industries, operational knowledge and frontline experience feed directly into each production cycle. Every new batch reflects the latest lessons from active deployment. This feedback loop between the battlefield and the factory floor is what sets this partnership apart.   Image: Maximilian Wied, Co-Founder and CFO at ARX Robotics and Anton Skrypnyk, Executive Chairman at Roboneers. Maximilian Wied, Co-Founder and CFO at ARX Robotics, shared: “Every week without the right equipment costs lives. ARX Industries is our joint answer of our partners and us to that urgency, a production engine built to deliver battle-proven UGVs to Ukrainian forces faster and at a scale that makes a real difference on the ground.”   Anton Skrypnyk, Executive Chairman at Roboneers, says:   “Ukraine has proven that robots win battles. Now, ARX Robotics and Roboneers are bringing that power to scale - delivering battle-proven UGVs to Ukrainian soldiers faster than ever. This is what 'Build with Ukraine' looks like in practice.” ARX Robotics launched an office this week in Berlin, to complement its new manufacturing facility in Munich and growing operations in London and Kyiv.

Danish martech startup Serpier has raised €1.4 million in funding from True Collective and the Export and Investment Fund of Denmark (EIFO) to expand its AI-powered marketing platform for e-commerce businesses. Founded in Aarhus in 2024 by Steffen Sørensen, Simon Holm, Søren Fuhr and Thomas Grástein, Serpier helps online retailers improve their visibility across both traditional search engines and AI chatbots such as ChatGPT and Gemini, as consumers increasingly rely on large language models for product discovery and recommendations. Its first AI agent, Navi, already manages visibility optimisation end to end by identifying opportunities to improve a customer's online presence, creating content and publishing it. The company plans to extend Navi's capabilities to include building landing pages, running marketing campaigns and automating additional marketing workflows. We've proven that our platform can create visibility for e-commerce businesses across both Google and AI chatbots like ChatGPT. Now we want to let our own AI agent Navi take on more of the marketing work itself — analysis, content and advertising — so marketing teams can focus on steering the direction instead, said co-founder Søren Fuhr, adding that Serpier sees marketing evolving towards a model where AI agents handle analysis and execution, allowing marketers to focus on strategy, prioritisation and key decision-making. The company says it generated more than €2.5 million in revenue in its first financial year and reached profitability. Serpier will use the new funding to develop its platform into an AI-powered marketing workspace where autonomous agents can automate a broader range of marketing tasks.