Hyperscale cloud and AI data centre developer and operator Pure Data Centres Group, together with AVK, a provider of prime, standby and dispatchable power solutions for data centres and AI infrastructure, today announced the launch of Europe’s first, large-scale, 110 MW on-site microgrid, developed to support early‑phase site operational resilience.

Located within Pure DC’s Dublin campus, the on‑site energy system provides dispatchable capacity to support data centre operations during the initial development phases, prior to full integration with the national electricity system, as grid connection capacity becomes available.

Over time, the campus is intended to operate as part of a hybrid energy configuration, combining grid‑supplied electricity with on‑site infrastructure designed to enhance flexibility, resilience and system stability.

While several microgrids are already in operation in the US, none have been in Europe until today. The deployment showcases the ability to use AVK’s microgrid technology for on-site power generation, and the transitional and complementary role it can play in supporting the delivery of strategically important digital infrastructure. This is particularly true in regions where grid reinforcement and renewable generation are being delivered on a phased basis under national planning frameworks.

Pure DC’s microgrid consists of three interconnected energy centres, with each building generating up to 30MW of power. Energy Centre 1 (EC1) and EC2 will be fully operational by the end of 2026, with EC3 to follow at a later stage.

The design includes Combined Heat and Power (CHP) capability, with infrastructure in place to enable heat recovery and potential future connection to district heating networks, subject to third‑party demand and regulatory approvals.

Waste heat recovery systems are also used to improve operational efficiency within the energy centres. Future water management measures include rainwater harvesting and on‑site treatment, reducing reliance on mains water for engine‑related processes.

The system is engineered to accommodate incremental changes in fuel composition, including hydrogen blending, supporting future decarbonisation of the gas network in line with national policy developments. Pure DC’s Battery Energy Storage System (BESS)  is integrated to manage load fluctuations and enhance operational efficiency, improving response times and enabling more optimal engine operation. The BESS is designed to support future renewable energy integration as part of a broader transition pathway.

Pure DC’s Executive Chairman and interim CEO, Gary Wojtaszek, said:

“The biggest barrier to deploying AI infrastructure in Europe today isn’t technology — it’s power. This microgrid proves that even the most constrained markets can unlock new digital capacity, giving Ireland the opportunity to lead Europe’s next chapter of AI infrastructure.

The future of AI infrastructure will be built where energy and compute come together — and that’s exactly what we’re building at Pure.” 

According to Ben Pritchard, CEO, AVK-SEG:

“This project demonstrates how carefully designed onsite energy infrastructure can complement national energy planning frameworks. 

This recognises that power is now the new differentiator for data centres, and that energy has shifted from being a utility to a strategic asset – shaping the location, design, economics and competitiveness for operators.

The first of many in Europe, this microgrid has the capability to revolutionise the data centre power race as we know it – providing a complementary solution that will ease gridlock and pave the way for greater take-up of AI and cloud.”

The company removed 159 million scam ads last year and took down 10.9 million accounts linked to criminal networks. Now it wants to catch scammers before they get to you. Meta has announced a fresh wave of anti-scam tools across its platforms, WhatsApp, Messenger, and Facebook, as it steps up both on-platform detection and cooperation […]

Spotlight Pathology, a UK healthtech company, has raised  £1.4 million in seed investment to support its development of AI software that analyses digital pathology images to support clinicians in identifying blood cancers faster and consistently.

Blood cancers are among the hardest to diagnose, often requiring multiple reviews by specialist pathologists. Delays can have serious consequences for patients, yet pathology departments across the UK are facing growing demand and a shortage of trained staff.

Designed to slot into existing clinical workflows, Spotlight’s technology helps pathologists prioritise cases and reach decisions sooner - enabling patients to begin treatment earlier.

Spotlight Pathology was founded by Dr Richard Byers, a Consultant Haematopathologist, and Dr Martin Fergie, an AI specialist with more than 15 years’ experience developing advanced algorithms for healthcare applications. The investment will support the company as it gains additional regulatory approvals and progresses through the first clinical in-use trials.

According to Sam Perona, Chief Executive Officer of Spotlight Pathology, blood cancers can be extremely challenging to diagnose, and diagnostic delays can have devastating consequences for patients:

“Our mission is to support pathologists with tools that fit seamlessly into existing workflows, helping them reach accurate diagnoses faster and with greater confidence.

This investment gives us the momentum to move from development into real-world clinical settings. We’re excited to be scaling the business from Daresbury, working alongside partners across the North West and beyond, and to be strengthening our board with experienced leadership as we enter this next phase.”

Sakura Holloway, Investment Director at the UK Innovation and Science Seed Fund, managed by Future Planet Capital, added:

“Spotlight Pathology is a strong example of how UK university research can be translated into technologies that will transform patient outcomes. The team are addressing a critical challenge in blood cancer diagnosis, and with the right leadership and support in place, the company is well positioned to bring this technology into clinical settings.

This investment reflects our conviction in both the team and the technology, and our commitment to backing spin-outs that apply advanced science to pressing healthcare challenges, improving productivity in the health system and delivering meaningful global patient impact."

Lead image: Richard Byers, Sam Perona, and Martin Fergie. Photo: uncredited.

Revolut has been awarded a full UK banking licence, after regulators lifted restrictions on the UK challenger bank, which had lasted for an extended time.

Revolut, valued at $75bn, today said it had received regulatory approval from the Bank of England's Prudential Regulation Authority (PRA) to exit the mobilisation phase, and launch as a bank in the UK.  

Nik Storonsky, co-founder and CEO of Revolut, said: “Launching our UK bank has been a long-term strategic priority for Revolut, and marks a significant moment in our journey.   

"The UK is our home market and central to our growth. We look forward to introducing a full suite of banking services to our millions of UK customers, bringing the same innovative experience we already provide across the rest of Europe. This is a vital step in our mission to build the world’s first truly global bank.”  

The winning of the licence draws to a close a 20 month process in which Revolut has been awaiting to get the full green light, after securing a licence with restrictions in July 2024. The period of restrictions usually lasts around 12 months.

In this so-called “mobilisation phase” Revolut has been operating under banking restrictions, including a cap on deposits. Revolut applied for a UK banking licence in 2021.

The licence win means that Revolut, which has 13m customers in the UK, will be able to begin offering accounts as a fully licenced bank in the UK for both retail and business customers. It enables Revolut to offer deposit accounts protected by the FSCS (Financial Services Compensation Scheme) on eligible deposits and paves the way for a wider range of services in the future, including lending and other products.  

It will allow it to better compete with UK established banks like HSBC, Lloyds, and Barclays and, given that a UK banking licence is held in high regard, could help with other licence wins around the world.  

Francesca Carlesi, UK CEO at Revolut, commented: “Becoming a bank in our home market marks a defining moment in our journey — a milestone achieved through relentless focus, discipline, and belief in what we’re building.   

"Securing this licence lays the foundation for our next chapter: expanding into a broader suite of products, including credit, to sit alongside the innovative services our customers already rely on every day. This will now enable us to continue on our mission to deliver the most seamless, secure, and customer-centric banking experience for consumers across the UK.”

Another Earth, a company developing AI-powered simulation and synthetic data for Earth observation, has raised a total of €3.5 million in funding. The round includes new investment from Wake-Up Capital alongside existing investors Rockstart, Inovexus, and Stamco AG, as well as support from the Austrian Research Promotion Agency (FFG) and Austria Wirtschaftsservice (AWS).

Based in Vienna, Another Earth develops technology that generates synthetic satellite imagery and geospatial datasets using generative AI and 3D modelling. The platform enables organisations to train and test AI models for monitoring environmental change and analysing land, water, and infrastructure at scale.

The company’s technology is designed to address a key challenge in Earth observation AI: limited access to high-quality training data. Traditional satellite imagery can be costly to obtain, particularly in remote regions, and preparing datasets often requires extensive manual labelling.

By generating synthetic satellite data from scratch, Another Earth can automatically produce labelled and segmented datasets, enabling organisations to train AI models more efficiently while reducing the cost and potential bias associated with traditional data sources.

Maya Pindeus, CEO and co-founder of Another Earth, said the planet is facing increasing challenges, including land degradation and climate-related disasters, and noted that artificial intelligence can help address these issues if it has access to the appropriate data.

The biggest barrier to scaling Earth Observation AI is the scarcity and prohibitive cost of high-quality training data. With this funding, and our deployment into vital ecosystems spanning from Latin America to Africa, we are generating data where there is none. We are giving organisations the tools to transition from reactive crisis response to proactive, predictive intervention.

Another Earth’s international expansion builds on its work in Sub-Saharan Africa, where its technology is used with GeoTerra Image to monitor the environmental impact of mining and industrial sites. The company is also expanding its Synthetic Data Platform in Brazil through a partnership with NovaTerra, focusing on applications such as deforestation monitoring, agricultural analysis, and climate-related risk assessment.

The new funding will be used to accelerate the deployment of the company’s Synthetic Data Engine and expand its use in environmental monitoring and risk simulation.

In particular, Another Earth plans to focus on applications across Brazil and Sub-Saharan Africa, generating high-resolution synthetic satellite data to support biodiversity monitoring, deforestation tracking, and environmental risk analysis in vulnerable ecosystems.

The EU–US transatlantic relationship has entered a new phase of volatility after the US Supreme Court struck down President Trump’s global tariff deals. The ruling casts fresh doubt over the 15% tariff rate on EU exports negotiated last summer, prompting the European Parliament to freeze the ratification of the new EU–US Trade Agreement pending legal clarity. In response to the court’s ruling, Donald Trump initially threatened to raise global tariffs to 15%, but instead imposed a temporary 10% rate set to expire on 24 July unless Congress renews it.

For Brussels, the implications go far beyond tariffs. With EU–US trade surpassing €1.6 trillion annually, rising legal uncertainty and renewed tariff threats amid geopolitical crises have strained not only economic ties but also the broader traditional alliance.

Implications for European startups

Amid this geoeconomic upheaval, and with President Trump previously threatening to impose additional tariffs on several European countries in his bid to take Greenland, how has the new EU–US trade deal and the ongoing uncertainty impacted transatlantic trade and, specifically, European startups?

Since July 2025, EU exports to the US have taken a hit, weighed down by higher US tariffs and a stronger euro, making European goods more expensive for US consumers. UN Comtrade data shows that in the third quarter of 2025, EU exports to the US fell sharply by 25% to €147.1 billion compared to the previous quarter. That figure had been temporarily inflated by a March surge, as US firms rushed to import European goods ahead of new tariffs taking effect on 1 August. The overall impact of the tariff agreement is stark: the EU’s goods surplus with the US has almost halved, dropping from €81 billion in the first quarter of 2025 to €41 billion in the third quarter.

Irish and European exporters to the US are impacted by the tariffs in several ways, with startups in trade-dependent sectors, including hardware, machinery, chemicals and pharmaceuticals, being particularly vulnerable. Tariffs increase costs, squeezing profit margins and disrupting supply chains, which in turn can impact cash flow, as Enterprise Ireland has warned. Rising prices and uncertainty may prompt US buyers to seek alternative suppliers, leaving European companies at a disadvantage compared to non-tariffed competitors. In response, Enterprise Ireland has introduced targeted grants for Irish exporters, both for market research and new market validation, to help them adapt strategically and diversify their markets.

Trade tensions have also revealed that access to critical technologies is no longer a given. The EU’s growing AI market, characterised by a proliferation of startups, is heavily dependent on US hyperscalers for cloud and AI services, representing a structural vulnerability. Moreover, the EU has faced an aggressive response as it has sought to regulate big tech platforms under the EU’s Digital Services Act, the Digital Markets Act and the AI Act. This has resulted in additional US tariff threats, with European regulators and service providers, including tech companies, also being targeted.

A climate of investment uncertainty

This volatile environment creates uncertainty, leading venture capital (VC) funds to make more cautious decisions, delay investments or even withdraw from Europe. In the first quarter of 2025, the US share of total European deal value fell to 46.9%, down nearly 4% from the previous quarter, according to PitchBook data. The overall volume of VC deals in Europe dropped to its second-lowest level in a decade in 2025, according to KPMG.

This particularly affects startups in less mature ecosystems that rely on external capital, especially from the US, to close funding gaps, often in the later financing stages. With the US already investing at least four times more VC per capita than Europe, there is a real risk of a widening innovation gap.

Against this backdrop, initiatives for greater harmonisation across EU member states are critical to alleviating the impact of trade tensions and tariffs. The Capital Markets Union aims to expand financing options for European startups and make the EU a safer and more attractive place to invest in the long term. Meanwhile, the €5 billion Scaleup Europe Fund, expected to launch in spring 2026, seeks to address shortages in late-stage growth capital for strategic deep-tech companies.

Transatlantic relationship no longer assured

The transatlantic relationship remains economically vital, but this once close relationship can no longer be taken as a given. This is a stark and painful lesson for Europe. Diversification is sensible, but it comes with its own political and economic trade-offs, and this was acutely reflected in the Mercosur Partnership Agreement negotiations.

A thriving startup ecosystem is crucial to Europe’s economic growth, competitiveness and resilience to global trade shocks. In its Startup and Scaleup Strategy from May 2025, the European Commission set out its role, together with the European External Action Service and Member States, in supporting startups to access global markets through trade agreements, EU Delegations and dedicated EU tools such as the Access2Markets portal. Innovative startup solutions also need to be integrated into the value chains of Global Gateway.

The moment is ripe for startups to consider the opportunities of diversification and how they can gain a strategic edge in Europe.

As Ireland prepares to host its eighth Presidency of the Council of the EU in the latter half of this year, trade relations are set to dominate the agenda. Given Ireland’s unique relationship with the US, it will have to continue to balance strained US relations within a broader European agenda as it assumes EU leadership. If Ireland’s approach is politically shrewd, it may well achieve this for the benefit of all Europeans.

The post Trade tensions and tariffs: What they mean for Europe’s startups appeared first on EU-Startups.

VC firm Samaipata has launched its third fund – Samaipata III – a €110 million vehicle aimed at backing early-stage tech startups building on the AI wave.

Samaipata plans to invest in 25 to 30 early-stage companies, with the capacity to deploy up to €10 million per startup over time.

The main focus is on AI-native businesses developing application-layer products that can scale internationally from day one.

The fundraising process is already well advanced, reaching €70 million. Institutional anchor investors include Spain’s SETT (Spanish Society for Technological Transformation) and Germany’s KfW, as well as several prominent Spanish family offices.  The fund also includes, as investors, founders who were backed by Samaipata in its first two funds and are now reinvesting in the firm as their companies have grown. 

Samaipata III will continue to capitalise on the firm’s Founder Success platform, designed to accelerate portfolio growth beyond capital alone. 

Founders gain access to a network of Operating Partners with experience at companies such as Anthropic, Google, Airbnb, Spotify and N26, who bring strategic perspective and hands-on operational expertise at key stages of development. The firm also facilitates introductions to potential clients and talent, while leveraging partnerships with leading technology players, including Nvidia, Anthropic, Microsoft Azure and Google Gemini, to strengthen technical capabilities and commercial traction

. Samaipata III will back projects that abstract the complexity of AI deployment for real-world use cases, primarily in B2B environments. 

“Samaipata III is launching at a particularly relevant moment for the European tech ecosystem. AI is moving beyond the experimental phase and beginning to integrate into critical processes with tangible impact. We see a clear opportunity to invest in teams capable of applying this technology in complex markets and building globally relevant companies from Europe,” José del Barrio, founding partner at Samaipata.

With an established European track record, Fund III builds on more than 44 investments across Spain and other key European markets, including the UK, France and Germany.

Samaipata’s early-stage portfolio stands out, with 80 per cent of Fund I companies advancing to Series A and 60 per cent of Fund II companies reaching that stage within five years, backed by leading international later-stage venture capital firms such as Accel, Creandum and Index Ventures. 

The portfolio includes companies such as Matera, Bigblue, Nory, Embat, VIVLA and Imperia. Deporvillage remains one of the firm’s most notable exits, sold to JD sports and achieving a 25x valuation increase from first ticket to exit. 

“Samaipata understood the business from day one and brought strategic judgment at key moments. Beyond capital, their involvement helped us execute with greater confidence as we scaled and ultimately supported the sale of the company. That experience also led me to invest in the fund myself after seeing firsthand how they work with founders.” Xavier Pladellorens, co-founder of Deporvillage and investor in all three Samaipata funds.

Sybilion has closed a $4.2 million seed funding round to develop what it describes as a decision platform designed to help industrial companies respond earlier to market changes and manage margin exposure in volatile conditions. The round was co-led by Venturefriends and Semapa Next and follows the company’s $600,000 pre-seed round, announced a few months earlier, which was co-led by Vanagon Ventures and EWOR.

Many manufacturers have access to historical data feeds, analyst reports, and internal forecasts, yet still find it difficult to determine which risk factors are most relevant for their operations at a given moment. Procurement, sales, and finance teams often rely on different data sources and reach different conclusions, and by the time decisions are aligned, market conditions may already have shifted, affecting margins. Even small timing discrepancies can have significant financial implications for companies operating with large cost bases.

Sybilion aims to address this challenge by analysing external market signals and linking them directly to a company’s cost structures and product portfolios. Rather than delivering standalone forecasts, the platform is designed to support decision-making by outlining potential options, trade-offs, and associated risk boundaries.

Dr. Bjol R. Frenkenberger, CEO and co-founder of Sybilion, noted that industrial companies typically have extensive data available but often lack clarity about which signals are most relevant and when decisions should be made.

Our goal is to give decision-makers the information advantage so they can turn external world dynamics into confident action before uncertainty becomes cost.

The system continuously processes a wide range of external indicators, including weather patterns, trade flows, freight rates, electricity futures, commodity prices, port congestion, industrial utilisation, and macroeconomic data, helping companies better understand the factors that may influence their operations.

Looking ahead, Sybilion plans to further develop its mapping between external signals and product-level exposure, expand integrations through its “Sybilion Connect” system so actions can be embedded directly into client workflows, and extend its platform from insight delivery toward agentic planning support that helps teams determine the next steps under uncertain conditions.

Swedish legal tech startup Legora has made its first acquisition, snapping up a Canadian legal AI startup, as it looks to expand its presence in North America. The acquisition for an undisclosed sum comes in the same week Swedish unicorn Legora announced its $550m Series D fund round, at a $5.5bn valuation.  

Legora, a much-hyped AI platform for lawyers which supports lawyers in researching, reviewing and drafting legal work, has acquired Walter.

Walter is a 10-strong team whose client roster includes law firms Fasken Martineau and McCarthy Tétrault. Walter bills itself as an “agent-native legal AI platform" for lawyers, which automates end-to-end legal workflows from email to finished document.  

Legora, founded in 2023, said the acquisition marks a significant step in Legora's push towards fully agentic AI workflows where its platform can carry out complex, multi-step legal tasks end-to-end, including document research, editing, and client replies.   

The deal will also help Legora expand its presence in Canada, as it looks to make its mark in North America, which it has earmarked as a key market.  

Legora already has offices in New York and Denver, with planned openings in Houston and Chicago.  

Max Junestrand, CEO and co-founder of Legora, said: “When we saw what the Walter team had built, we immediately recognised a shared philosophy around agent-native design. The Walter team have approached legal AI the same way we have – embedding closely with lawyers and designing agents to handle real, end-to-end workflows. Bringing our teams together allows us to scale that vision faster.”  

Ryan Wilson, co-founder and CEO Walter, said: “When we met the Legora team, it was clear we had a shared vision for the future of agentic legal technology.

“By joining Legora we can accelerate the realisation of that shared vision of end-to-end matter management with fleets of agents.

“We’ve built both companies in close partnership with the legal teams actually doing the work. Working with our customers, not just for them – iterating on feedback until the product matches how legal work actually gets done. And we’ve both arrived at the same conclusion: the future of legal AI is agentic."

Today, Porto-based Sybilion announced a €3.6 million ($4.2 million) Seed round to build what it calls a decision layer, designed to give industrial companies the ability to act earlier and protect margins in volatile markets.

The round was co-led by Venturefriends and Semapa Next. This announcement comes just months after their €517k ($600k) pre-Seed round, co-led by Vanagon Ventures and EWOR.

“Industrial companies do not lack data,” says Dr Bjol R. Frenkenberger, CEO and co-founder of Sybilion. “They lack clarity about which signals truly matter and when to commit. Our goal is to give decision-makers the information advantage so they can turn external world dynamics into confident action before uncertainty becomes cost.”

Sybilion’s Seed round comes amid a broader wave of European investment in AI-driven platforms designed to improve procurement, supply chain visibility and operational decision-making.

In 2025, Magentic, based in London, raised €4.6 million to scale autonomous AI agents that automate procurement and supply-chain operations, while fellow London startup Procure AI secured €11 million to build AI-native tools aimed at helping enterprises stabilise sourcing and supply chains. Another UK company, Monq, raised €2.5 million in pre-Seed funding to develop an AI-driven strategic negotiation platform designed to improve procurement outcomes.

Elsewhere in Europe, Berlin-based Freshflow secured €6.5 million to expand its AI platform for automated ordering and inventory optimisation in grocery supply chains, while London-based Treefera raised €26.2 million in a Series B round to scale its supply-chain transparency and risk intelligence platform.

More recently, Berlin startup Andercore secured €33.5 million in equity and debt financing to expand its AI-enabled industrial trade platform across Europe.

Together, these financings represent roughly €84 million invested across adjacent segments of industrial decision intelligence and supply-chain software over the past year.

Within this landscape, Sybilion’s approach – linking external signals such as commodity markets, logistics data and macroeconomic indicators directly to procurement, pricing and operational decisions – reflects a growing focus on AI systems that move beyond forecasting dashboards toward structuring and guiding real-time business decisions in volatile industrial environments.

“Industrial companies are being forced to make larger decisions on shorter timelines as volatility becomes the norm. We’re excited to support Bjol and the team as they become the decision layer for manufacturing,” adds Apostolos Apostolakis, Founding Partner at VentureFriends.

Founded in 2021, Sybilion connects external world dynamics – commodity markets, energy, weather, logistics and macro signals – to the decisions that shape margin, working on top of systems companies already run.

According to the company, a few weeks of mistimed procurement in manufacturing can erase millions in margin. Yet most of the industry still makes billion-euro decisions using spreadsheets, fragmented analyst reports, and instinct.

Most manufacturers today have access to historical data feeds, analyst reports, and internal forecasts – but Sybilion believes they still struggle to answer which risk factors actually matter for a company, for a product set, at this moment.

Procurement, sales, and finance often work from different inputs and reach different conclusions. By the time alignment happens, markets have usually already moved and margins are diminished. This cost is detrimental; even a three to five percent timing error on a €172 millions ($200 million) cost base can reportedly translate into millions in margin erosion.

Sybilion approaches the problem from the outside in. The system identifies which signals materially affect a company’s exposure and links them directly to cost structures and product portfolios. Rather than delivering another isolated forecast number, Sybilion structures the decision moment itself, clarifying realistic options, trade-offs, and quantified risk boundaries so companies can commit earlier.

Its platform filters more than one trillion external risk factors, including weather anomalies, trade flows, freight rates, electricity futures, commodity prices, port congestion, industrial utilization, and macroeconomic indicators.

Grégoire Viat, Principal at Semapa Next says: “We were impressed by what Bjol and the team at Sybilion have built in a short period of time. Sybilion delivers clear, measurable value to industrial customers, addressing a fundamental need for decision confidence in an increasingly volatile supply chain environment. We are pleased to support the founders as a long-term partner as they continue to scale the business.”

Over the past twelve months, Sybilion has grown annual recurring revenue to high six-figures, with zero churn and no sales team.

Use cases:

• K.D. Feddersen, an international distributor of engineering plastics, used Sybilion to align earlier on critical pricing and purchase decisions around global polymer trade flows and feedstock dynamics.
• At Jobachem, the platform enabled timely commitment with integrated energy futures and upstream commodity signals to frame procurement decisions with quantified risk boundaries.
• For Maral Overseas, forward trade flow analysis made it possible to take informed export allocation decisions by identifying regions where demand was strengthening.

Looking ahead, Sybilion plans to deepen its mapping from external signals to product-level exposure and decision recommendations, broaden “Sybilion Connect” integrations so actions land directly inside client workflows, and expand from insight delivery into agentic planning support that helps teams determine the next best move under uncertainty.

The long-term goal is to give industrial decision-makers an advantage that compounds, turning uncertainty from a threat into an edge.

“Bjol is one of the most fascinating founders I’ve been fortunate enough to meet. He is a piano prodigy, a builder, an academic, a leader,” says Daniel Dippold, CEO and Founder of EWOR. “With Sybilion, he managed to build the largest dataset of time-series data I have seen to date and orchestrates it in a way that gives industrial teams a decision advantage no one else can offer. Similar to Bjol, Sybilion is one of a kind. Their technology is built like a musical masterpiece, they are growing fast, and Bjol has assembled a unique team that would have never come together were it not for his leadership.”

The post Portuguese startup Sybilion secures €3.6 million to build AI-powered decision layer for industrial companies appeared first on EU-Startups.

Living Models, a Paris–Berkeley startup, has raised $7 million in seed funding as it emerges from stealth to develop foundation models for biology trained on DNA, RNA, and multi-omics data aimed at improving understanding of biological systems.

To support the next stage of development, the company has also secured access to a computing cluster of 120 NVIDIA B200 GPUs, which it plans to use to train its next generation of biological AI models.

The company develops large-scale transformer models trained on genomic, transcriptomic, and other biological datasets to analyse patterns within living organisms. Operating in Paris and Berkeley, Living Models brings together researchers in artificial intelligence and plant science to apply machine learning to biological research and agricultural innovation.

While artificial intelligence has already transformed sectors such as finance, software development, and content creation, its application in areas such as agriculture and food production remains at an earlier stage.

Living Models is focusing on this area by applying AI techniques to biological data, particularly in plant science, where improving crop resilience and productivity is becoming increasingly important as climate pressures affect global agriculture.

As part of its launch, the company introduced BOTANIC, a family of transformer models designed for plant biology. The models are trained on genomic sequences from multiple plant species and analyse genomic and other biological data to identify genetic markers associated with traits such as climate resilience and disease resistance.

By predicting which genetic variants are worth testing, the technology aims to help seed and agricultural companies accelerate the development of new crop varieties.

OpenAI trains on Reddit and Wikipedia to understand human language. We train on DNA, RNA, and gene expression to understand the language of life itself,

said Cyril Véran, CEO and co-founder of Living Models.

Traditional crop breeding cycles can take many years, partly due to the time required to identify promising genetic traits. By analysing genomic data computationally, Living Models aims to shorten the early stages of this process by helping researchers focus on the most relevant genetic variants before conducting field validation.

In the longer term, Living Models plans to expand its work on foundation models for biological systems beyond plants. The company began with plant biology due to the availability of large genomic datasets, faster validation cycles compared with other life-science fields, and the growing need for technologies that support climate-resilient agriculture.

On 7-8 May, Malta will become the meeting point for some of Europe’s most ambitious founders and investors as we host the 12th edition of the EU-Startups Summit. The event will bring together around 2,500 founders, startup enthusiasts, corporates, angel investors, VCs, and media representatives from across Europe, all with a shared interest in startups with global ambitions.

This year’s agenda will feature insights from some of Europe’s most successful entrepreneurs and explore the founder journey in depth. Across two stages, attendees can expect inspiring talks, valuable networking opportunities, hands-on workshops, and a buzzing exhibition hall. For the first time, the Summit will also host the Global Startup Awards Grand Finale, bringing the world’s largest independent startup ecosystem platform into the programme for a special celebration of leading founders, startups, investors, and ecosystem players from across the globe.

A global competition reaches its final stage

Following the conclusion of regional competitions across six continents, the Global Startup Awards has announced the global finalists for its 11th edition. After processing more than 71,000 nominations, collecting over 403,000 public votes, and completing jury evaluations across its international network, the organisation has now confirmed the founders, startups, investors, and ecosystem leaders advancing to the global stage.

Operating across 154 countries, the Global Startup Awards works with regional partners, governments, ecosystem leaders, and industry experts to identify standout innovators and contributors from startup ecosystems around the world. This year’s finalists come from Africa, Asia, Europe, the Middle East, North and South America, and Oceania, reflecting the breadth and diversity of global entrepreneurship today.

Kim Balle, CEO and Co-founder of XO Group, said: “After more than a decade expanding the GSA to 154 countries worldwide, we are now refocusing on Europe. Our goal is to strengthen the European startup ecosystem by creating innovation bridges to the rest of the world through our global platform.”

Competing for 13 Global Titles

The global finalists will compete for 13 prestigious titles during the Global Grand Finale. The awards recognise both startups and key ecosystem contributors across the following categories:

Startup of the Year
Best Newcomer
Best Greentech Startup
Best Health Tech Startup
Best Fintech Startup
Best Youth Startup
Investor of the Year
Best Accelerator or Incubator Program
Best Co-working Space
Founder of the Year
Diversity Role Model of the Year
Ecosystem Hero of the Year

Unlike the other categories, the People’s Choice title is open to all global finalists, regardless of their primary category, and is determined entirely by public votes. The full list of global finalists for edition 11 is available here

The road to the global grand finale!

This year’s edition also marks a notable first for the organisation. For the first time, the Global Grand Finale will be held as part of the EU-Startups Summit, bringing the awards to one of Europe’s leading annual gatherings for founders, investors, and startup ecosystem players.

The collaboration has been made possible through MeOut Group, the investor behind the Global Startup Awards and parent company of EU-Startups. As part of the Summit programme, finalists will be showcased in front of an international audience before the global winners are announced during the awards ceremony.

By integrating the Global Startup Awards into the EU-Startups Summit, edition 11 creates a stronger link between GSA’s worldwide network and the European startup scene, while opening the door to new international connections and collaborations.

Kim Balle added: “Europe has all the ingredients to lead in this new era of innovation. What we need to see is less friction and stronger alignment between founders, capital, and policy so that innovation can scale faster across the continent. The Global Startup Awards Grand Finale days, as part of the EU-Startups Summit, will not only celebrate the achievements of our finalists. They will also bring together innovation leaders from across Europe and help build innovation bridges between the European ecosystem and the rest of the world.”

Tickets and access

As the Global Grand Finale will take place within the EU-Startups Summit, attendees are required to hold a Summit ticket. Global finalists will receive one complimentary ticket, while other attendees can use the discount code GSA25 to receive 25% off their ticket purchase.

The post Global Startup Awards announces global finalists ahead of historic Grand Finale at the EU-Startups Summit! appeared first on EU-Startups.

Founded just months ago by two repeat Y Combinator founders and a retail veteran, Lemrock is betting that AI agents are becoming the new storefront,  and that brands are nowhere near ready for it. When you ask ChatGPT or Perplexity which running shoes to buy, someone has to make sure the right brands show up, […]

For years, manufacturers have experimented with Industry 4.0 — running pilots in predictive maintenance, production monitoring, and AI-driven optimisation.

Yet many of these initiatives struggled to move beyond proof-of-concept. The underlying problem wasn’t a lack of ideas, but the difficulty of accessing and structuring machine data across complex factory environments.

United Manufacturing Hub Systems (UMH) is building an open-source data platform designed to solve exactly that challenge. The company now counts manufacturing, food and beverage, and top-20 automotive suppliers among its customers, supporting deployments across more than 150 sites globally.

Oh, and the company raised €5 million in January. 

I spoke to Nikklas Hebborn, CCO of UMH, to learn more.

 From venture investor to operator

Before joining UMH operationally, Hebborn was a partner at Freigeist Capital, the German deep-tech venture firm founded by Frank Thelen. His background spans venture capital, consulting, and advisory roles with organisations including Bayer, Capgemini Invent, Roland Berger, and Kearney.

Hebborn later transitioned from investor to operator, moving from backing United Manufacturing Hub as an early investor to joining the company’s leadership team.

Founders who experienced the problem firsthand

The company was founded by CEO Alexander Krüger and CTO Jeremy Theocharis. After graduating from RWTH Aachen University, the pair worked on digitalisation projects for large consultancies such as McKinsey, travelling globally — from Tokyo and Singapore to Atlanta — deploying industrial use cases directly on factory shop floors.

Over several years of doing this work, they noticed a recurring challenge. Building the actual use case — whether a dashboard for energy monitoring, productivity tracking, or even deploying AI — represented only about 10 per cent of the work. The remaining 90 per cent involved collecting and preparing the data: gathering it in real time, ensuring it had the correct format and context, and maintaining sufficient quality.

“That’s where most projects struggled,” explained Nikklas Hebborn.

The real bottleneck in industrial digitalisation

The founders realised that the real bottleneck wasn’t the applications themselves but the infrastructure required to reliably access industrial data. They therefore focused on building the underlying layer that connects operational technology (OT) — machines and sensors on the factory floor — with IT systems such as ERP platforms.

The result was a platform built around what the company calls a Unified Namespace: a structured data environment that allows companies to move seamlessly from a site-level overview down to individual machines or even specific sensor readings.

According to Hebborn, the founders’ deep industry experience was critical to shaping this approach.

“Many startups identify a problem and then bring in domain expertise later,” he said.

“Here it was the opposite — the founders lived with the problem for nearly a decade.

They had personally experienced the pressure of delivering digitalisation projects under tight timelines while working on shop floors around the world. That deep understanding of the problem space was what convinced me to join.”

UMH’s platform helps manufacturers collect and structure data from machines, sensors, and factory software systems. Modern factories run a mix of legacy equipment, industrial controllers, and enterprise software, all producing data in different formats. The platform gathers machine data via common industrial protocols and transforms it into a unified, real-time stream that feeds dashboards, analytics tools, manufacturing execution systems (MES), or AI models.

A key concept behind the platform is the “Unified Namespace,” which acts as a single source of truth for factory data. Instead of each application pulling information separately from machines or databases, data is published once into a shared structure that authorised systems can access. This simplifies integration, improves transparency across production processes, and accelerates Industry 4.0 use cases such as predictive maintenance, energy optimisation, and production monitoring.

Under the hood, the platform is designed as a modular infrastructure layer with tools to manage deployments across factories. UMH’s solution has two main components.

The first is the infrastructure layer, configured through code via a large configuration file. On top sits a management console that acts as a control centre for deploying instances, connecting machines, building data bridges between systems, and defining data models. Hebborn explained:

“One automotive supplier we work with has CNC machines worldwide. Each machine produces millions of data points, but only a handful are actually relevant. The raw outputs are often cryptic values like ‘DB3456’, which might represent temperature or pressure.”

Within the console, companies create data models that translate these signals into contextualised, understandable information.

The platform supports two interaction modes. Developers can configure deployments through YAML files — something Hebborn says AI tools can generate quickly when connecting hundreds of machines. For non-technical users, UMH also offers a visual drag-and-drop interface, which becomes important when deployments scale across dozens of sites.

The value of open source and interoperability 

A defining decision of UMH was to build the platform as open source, an unusual move in an industry dominated by large incumbents.

“There are many big players here, from Siemens and Rockwell to the hyperscalers on the IT side trying to move industrial data into the cloud,” said Hebborn.

“But customers increasingly want to own their software. If you want to stay competitive and become something like a Tesla in manufacturing, you need to be software-driven and vertically integrated. Open source allows companies to maintain that ownership.”

The second reason, he explained, is interoperability. “

A typical factory floor might use Siemens machinery, Rockwell automation, and Microsoft cloud infrastructure. Companies need an independent layer in the middle that connects all of these ecosystems.

To achieve that, the founders built its platform around an open-source model and cultivated a broad community around it. Today, more than 1,000 system integrators, consultants, and end users are using the community edition.

“That effectively gives us hundreds of people constantly testing the product, identifying issues, and contributing feedback,” Hebborn said.

“Many competitors develop a product internally and only receive feedback once it’s deployed at a customer. Our development cycle benefits from a much larger real-world testing base.”

That community-driven approach has also helped drive organic adoption. “Interestingly, many customers actually discover us themselves,” he added.

“Often, they first attempt to build their own solution internally. Eventually they realise how difficult it is to scale and then start looking for an infrastructure layer like ours.”

Why the real challenge wasn’t the application

Industry 4.0 and Industrial IoT have been discussed for years, but many companies have run pilots that have struggled to scale, and no dominant platform has emerged. According to Hebborn, what has changed over the past five to ten years is the availability and accessibility of machine data.

“Ten years ago, getting data from industrial machines was difficult. Vendors like Siemens built closed ecosystems.

Today, there are far more standardised protocols and APIs available to extract that data. Another shift is infrastructure.

Production sites now have much better IT connectivity, including direct internet access on the shop floor. And finally, companies have already experimented with digitalisation. 

Many consultancies sold digital transformation projects that led to numerous pilots and proof-of-concept use cases. Many of those didn’t scale, but the demand didn’t disappear.”

Today, Hebborn sees companies with a“chessboard” of use cases they want to implement.  “They know the potential is there—they just lacked the underlying infrastructure layer to do it properly.”

A  key decision for the company was focus. Its CTO has a strong opinion about this: we want to be the best data layer, not the best tool for everything, explained Hebborn.

“For example, we don’t try to build our own visualisation tools. Customers can use platforms like Grafana, Power BI, or Snowflake.

We simply ensure the data is structured and accessible so those tools can use it. If we tried to build every component ourselves—  visualisation, historians, analytics — the product would become too complex and we would risk turning into a consultancy.”

Hebborn stresses that you can’t bluff in a factory environment. 

“One example was HiPP, the infant nutrition company. When we first installed the system and showed them the dashboard, the management team asked, “Is this really our data on the screen?” They had never seen their production data presented in that way before."

He found this surprising.

"We weren’t building something as complex as an electric vehicle, we were simply connecting data that already existed on the shop floor. But clearly, this problem still hasn't been solved properly.”

Another important element for UMH is training the customer team. It follows a “train-the-trainer” model where it trains one production specialist who then trains colleagues across the site and other facilities. Hebborn shared:

“In one case we didn’t hear from the team for two or three months. But we could see them continuously developing new use cases internally — and eventually they expanded the deployment to additional sites.”

Hebborn is modest about the company's success, sharing that while the company may not have hundreds of customers, "every customer we do have has expanded their deployment — and they’ve typically done so within twelve months."

"In manufacturing, that is extremely fast. If we eventually reach three or four hundred companies and scale across their sites, that would already represent a very large business. We don’t need thousands of customers to build a major company. That’s the proof point we’re most proud of today.”

UMH's next phase is about scaling through team expansion. Geographically, the team is not aggressively pushing international expansion yet as the DACH region already has a huge concentration of global manufacturing companies. 

“Many of them operate internationally, so once we deploy locally, the solution often spreads across their global sites,” shared Hebborn.