The most robust, reliable, and productive uses of AI are when many models are used in coordination with each other, with "regular" programming mixed in. This leads to more capable, more reliable, and more interpretable AI systems. Most people building with AI already know this. But (unless you work for Google) the main barrier to realizing multi-step AI workloads is an infrastructure one. For the rest of us, we’re left creating an unwieldy mess of chained API calls to multiple providers, with many roundtrips in between. This state of affairs is stifling progress. Nobody wants to deal with more tooling and infrastructure… but everyone would benefit from simple, intuitive interfaces that abstract away a powerful system underneath. Substrate is the first inference API optimized for multi-step AI workloads. With Substrate, you can write less code, run more inference, and build high performance AI applications with zero infrastructure to manage. No tooling, no infrastructure – just elegant abstractions. Excited to support Ben Guo, Rob Cheung and the amazing Substrate team on their mission to democratize how AI applications are built.
Last September, I left Stripe after 8+ years to build something new with my friend Rob Cheung. Today, we're launching Substrate, the first inference API optimized for multi-step AI workloads. https://substrate.run With Substrate, you connect nodes from a curated library that includes optimized ML models, built-in file and vector storage, a code interpreter, and logical control flow. By simply connecting nodes, you describe a graph workflow, which Substrate then analyzes and runs as fast as possible. Entire graphs of many nodes will often run on a single machine, with microsecond communication between tasks. With this launch, we're announcing our $8M Series Seed led by Lightspeed with participation from South Park Commons, Craft Ventures, Red Swan Ventures, and Lorimer Ventures, joined by industry luminaries like Guillermo Rauch, Immad Akhund, William Gaybrick, Chris Best, Gokul Rajaram, Shreyas Doshi, Jana Messerschmidt, Michael Manapat, and Matthew Hartman. We’ve been working on Substrate privately for nearly a year. We’ve battle-tested the product with great customers like Substack, and we’re finally ready to open access to everyone. Substrate was built by an exceptional team: Liam Griffiths, Rishabh Parikh, 💭 Kyle Pitzen, and Chris S., who bring years of experience building world-class software at Stripe, Pulumi, Robust Intelligence, Confluent, Zendesk, and Grailed. Truly, I have never enjoyed working with a group of people more in my entire career. Thank you to all our friends and early supporters for helping us in the first chapter of this adventure: Aditya Agarwal, Ruchi Sanghvi, Shreyans Bhansali, Namrata Patel, Andrew Kortina, iqram and the immigrant groove, Sam Lessin, Julian Connor, Mike Cohen, Jairaj Sethi, Chris Mueller, Stanislas Polu, Sabrina Hahn, Anne Lee Skates, Audrey Kim, Quinn Garner, Jason Cui.
Rob Cheung
5mo
honored to have you on board！
Mahesh Ram
5mo
This makes absolute sense. Smart idea...
Helping revenue teams humanize selling
5moSo this is a point lost on most. Engineers, VCs, sellers, and pretty much everyone except good ML engineers, data scientists or PMs. A model is not the moat, but the right collection and orchestration is. Why? Because it becomes a solution at that point. And the ability to build a solution using the right kind of models, fine-tuning, training, data, data cleaning and structuring, it all comes together to become a beautiful solution and an even stronger moat. When people say what's your data moat or what's your model moat, they basically don't know what they are talking about. Or in the best case, know a little bit.