Meticulous raises $15 million to test AI-generated code

Meticulous, a London startup building testing software for AI-generated code, has raised a $15 million Series A led by Ethan Kurzweil at Chemistry, with Menlo Ventures also participating. The company says the funding will go into marketing, R&D and hiring as development teams face a growing review burden from code written by AI agents.

The valuation was not disclosed. The round also included angel investors Lachy Groom, Poolside cofounder Jason Warner, Dropbox cofounder Arash Ferdowsi, former Adobe chief product officer Scott Belsky, Vercel founder Guillermo Rauch, former OpenAI engineering lead Calvin French-Owen, Hex CTO Caitlin Colgrove and former Cursor head of engineering Jason Ginsberg.

Meticulous was founded by brothers Gabriel Spencer-Harper and Quentin Spencer-Harper. Quentin previously spent more than a decade at Palantir, while Gabriel worked as a software engineer at Dropbox. The company is positioning itself around a practical problem in AI-assisted software development: code generation has accelerated faster than review and testing workflows.

How the product works

Meticulous analyzes a customer’s codebase and application, then builds a set of edge cases that could expose broken logic or crash risks. When a human developer or AI coding agent changes code, the system simulates user flows before and after the change and identifies the cases affected.

Before code is merged, developers receive a visual report showing sequences of thumbnails for affected scenarios. Quentin Spencer-Harper told Sifted that a report could show how a change alters a screen for different user types, such as premium and standard users, allowing developers to see the effect before shipping.

The company says developers still make the final call on releases. They can ship the change or ask the coding agent they use to revise it. Quentin Spencer-Harper also told Sifted that the system is deterministic, which lets it warn users and agents when relevant cases are not covered.

Meticulous currently tests frontend code only. The company plans to expand into backend and performance testing, which would put it closer to full-stack quality assurance at a time when AI coding tools are pushing more code through engineering organizations.

Customers and hiring plans

Meticulous says annual recurring revenue has increased fivefold over the past year, though it did not disclose current ARR, revenue retention or customer count. Its named customers include Notion, ElevenLabs, Dropbox, Wiz and LaunchDarkly.

Gabriel Spencer-Harper told Sifted the company has historically grown through word of mouth and now plans to spend part of the new capital on marketing. The shift suggests Meticulous is moving from founder-led adoption inside engineering teams toward a broader commercial push.

The company has 20 employees and wants to reach roughly 30 to 40 people over the next 12 months. It is also hiring forward deployed engineers, a role used by several AI companies to place technical staff close to customers and adapt products to specific deployment problems.

Gabriel Spencer-Harper told Sifted the market for engineering talent is competitive. For Meticulous, the hiring plan is still modest by venture-backed software standards, but the company is adding go-to-market and deployment capacity as AI coding tools create more demand for automated verification.

This story draws on original reporting from Sifted.