Latent Labs

🦠 On making biology programmable... Quite the episode of Nothing Ventured this week! I spoke to Simon Kohl, founder of Latent Labs, a London & SF-based startup using generative models for computational protein design. Simon studied physics at the Karlsruhe Institute of Technology before pivoting to machine learning for medical imaging during his PhD. His work led to an internship at Google DeepMind, where he later joined the AlphaFold 1 and 2 teams, developing key uncertainty metrics for protein structure predictions. He then co-led DeepMind’s generative protein design efforts, setting up a wet lab to validate AI-designed proteins. Seeing a gap in bringing this tech to market, he founded Latent Labs. Now 1.5 years in, Latent Labs has just raised a $50M Series A led by Sofinnova Partners and Radical Ventures to advance AI-driven biotech. We talked about: 🧬 Generative modelling and protein modelling. 🔬 Transforming science building AlphaFold 2 within Google Deepmind. 🥼 Moving biology from the lab on to the computer. 💊 Building frontier models to improve the process of drug discovery. 🇬🇧 The UK and Europe are powerhouses for scientific progress. 💬 The way you present an idea matters, be agile, but maintain your convinction. 💰 When fundraising, nimbleness should not translate to indecision. 🌎 Building teams in multiple locations is critical if you want to access the best talent. 📈 Simon’s controversial conviction: it’s accepted wisdom that platform companies can’t scale as well as others, but that simply isn’t the case. Join tens of thousands of venture curious individuals listening to Nothing Ventured to level up their understanding of all things venture, tech and finance ✨ Check out the Quickfire I did with Simon below and I'll drop links to the episodes in the comments 👇🏾👇🏾👇🏾

Great youtube video by Vertasium, suggesting AlphaFold 2 may be "The most useful thing AI has ever done". I recommend checking it out for some great background on the achievement and why its has been so significant for structural biology. It also features the moment when John told us about the AlphaFold 2 results in CASP14, see below. Let's just say I might have to work on my external display of excitement :) Link to full video: https://lnkd.in/ebSjb__c

"We are working on making biology programable." Latent Labs CEO Simon Kohl tells Tom Mackenzie how his firm uses artificial intelligence to reduce risk and speed up timelines in drug discovery. https://trib.al/4Sdn57y

This morning, I had the privilege of discussing Latent Labs' recent $50M funding and our mission to make biology programmable on Bloomberg News Daybreak Europe. Our goal at Latent Labs is to build frontier AI models that learn the fundamentals of biochemistry. Thank you to Tom Mackenzie and the Bloomberg Daybreak team for the conversation. For those interested, you can watch the full interview starting at 37:05 via the link below: https://lnkd.in/edCauK_q

Latent Labs comes out of stealth today with $50M funding. Our goal? To push the frontiers of generative biology, giving partners instant access to tools capable of accelerating drug design. Every biotech or pharma company searching for the best therapeutic molecules understands the role AI can play - but not all are in a position to develop their own advanced models. That’s where Latent Labs comes in. The funding includes our Series A co-led by Radical Ventures and Sofinnova Partners, with existing investors Pillar VC, Kindred Capital VC and 8VC participating alongside new investors Isomer Capital and Flying Fish Partners. Our angel investors include Google Chief Scientist Jeff Dean, Transformer architecture co-inventor and Cohere founder Aidan Gomez, and ElevenLabs founder Mati Staniszewski. I loved my time at DeepMind working on AlphaFold2 and to be there in Stockholm to watch Demis Hassabis and John Jumper accept the Nobel Prize was incredible. It is clear protein design has the potential to transform medicine and scientific discovery. I believe Latent Labs will be the team to fully realize its potential. It’s a vast field and there’s so much still to do. Let’s go! Apply to access our tools or join the team www.latentlabs.com.