Undo Raises $37M to Fix AI Code Chaos

Jan de Vries · 2026-06-15

Cambridge-based Undo just raised $37 million to solve a problem every engineering team is starting to feel: AI-generated code is getting messy, and nobody can debug it. ### The Funding Details The round was led by Elsewhere Partners, and it's all about giving AI agents the runtime context they need to actually understand what code does when it runs. Not just what it says on paper. "We're ahead of the curve," says Undo Founder and CEO Greg Law. "Undo has spent years building deterministic, program recording technology for code failure runtime visibility, which has become absolutely essential with the rise of AI." ### Why This Matters Now Here's the thing: AI models can generate code faster than ever, but that code is often poorly understood, poorly structured, and full of unknowns. Systems become unstable. Outages happen. Security breaches follow. Rod Favaron, Operating Partner at Elsewhere Partners, puts it bluntly: "AI is making code unmanageable – introducing code that engineers cannot understand, trust, or debug." Undo's technology captures the complete execution history of a program into self-contained recordings. That means AI agents can see exactly what the code did when it ran, not just what the code says. And that changes everything. ### The Numbers That Prove It Undo's benchmarks on complex bugs tell a compelling story: - The latest AI models can identify root causes just 38% of the time without Undo. With Undo's runtime context, that jumps to 92%. - When models can solve a problem without Undo, they use fewer tokens when they have Undo's context. - Customers report completing root-cause analysis 100 times faster with Undo. ### Enterprise Adoption Palo Alto Networks is already on board. Suresh Sangiah, Senior VP Engineering, says: "Quality is extremely important at Palo Alto Networks, and we cannot afford to rely on guesswork. The hardest – and costliest – bugs in multi-million-line codebases live in runtime state and are not captured by logs or other solutions." ### The Bigger Picture Undo sits at the intersection of several hot areas in AI-assisted software development: software supply chain control, AI-agent security and governance, CI/CD, cloud operations, production infrastructure, and developer tooling. Similar UK-based companies like Cloudsmith, Geordie AI, Trent AI, CodeWords, Overmind, and Toyo all raised money in 2026, showing this is a real trend. ### What's Next Founded in 2012, Undo is now embedding its technology directly into AI workflows. The goal? Make AI-generated code as reliable and maintainable as code written by humans. Because no matter how good AI models get, results are limited by how good the context is. AI = model + context. And Undo provides the context.