- London-based Meticulous has raised $15 million in a Series A led by Chemistry, with participation from Menlo Ventures and a roster of AI industry leaders.
- As coding agents generate software faster than engineers can review it, Meticulous automatically tests every code change before deployment.
- The company has grown its annual recurring revenue fivefold over the past year and now serves customers including Notion, ElevenLabs, Dropbox, Wiz and LaunchDarkly.
London-based Meticulous has raised $15 million in a Series A led by Chemistry, with Menlo Ventures and a stacked angel bench: Lachy Groom, Jason Warner, Arash Ferdowsi, Scott Belsky, Guillermo Rauch, Calvin French-Owen, Caitlin Colgrove and Jason Ginsberg among them.
The pitch is blunt: AI agents now write code faster than any team can review it, so Meticulous tests every change automatically before it ships.
Testing as the new chokepoint
According to Grand View Research, the global software testing market was worth an estimated $49.36 billion in 2025 and is projected to reach $93.15 billion by 2033. That’s the unglamorous half of a very glamorous trend: Tech Funding News has tracked a wave of funding into AI coding agents that write software faster than any human team can check it, from Blitzy’s $200 million raise to Cognition AI’s push toward a $25 billion valuation. Every one of those stories has the same subtext: somebody has to check the work.
Meticulous, founded in 2021, by brothers Gabriel Spencer-Harper and Quentin Spencer-Harper works by scanning an application’s codebase, finding the edge cases a given change could break, and simulating real user sessions against the update before and after. Developers get a visual diff instead of a wall of test output. Right now it’s frontend-only; backend and full-stack validation are next.
Gabriel Spencer-Harper spent time as a software engineer at Dropbox. His brother and co-founder, Quentin, spent a decade leading frontend engineering on Palantir’s Foundry product. Both say the same lesson kept surfacing at very different companies: you can test code exhaustively, and it still breaks the moment real users touch it.
Rather than relying on developers to write and maintain thousands of brittle automated tests, Meticulous continuously observes how users interact with an application and automatically generates comprehensive end-to-end tests that evolve alongside the product.
A crowded fight for the trust layer
Meticulous says its annual recurring revenue is up fivefold over the past year, with Notion, ElevenLabs, Dropbox, Wiz and LaunchDarkly among its customers.
It’s not the only one chasing this. Harness, valued above $3.7 billion, has pushed into AI-powered testing and delivery. Diffblue has raised over $50 million automating Java unit tests. CodeRabbit and Qodo are baking AI straight into code review.
Elsewhere in TFN’s coverage, SolveAI’s $50 million raise shows the same logic applied one layer up — not just generating code, but guaranteeing it’s production-ready. What Meticulous is betting on is narrower: not writing the code, not reviewing the pull request, but checking whether the thing actually works before it goes live.
Unlike many AI coding startups focused on generating software, Meticulous positions itself as the trust layer sitting between AI-generated code and production, allowing engineering teams to move faster without sacrificing reliability.
The new funds go toward R&D, backend and performance testing, and growing the team from around 20 people to somewhere between 30 and 40 over the next year. That brings Meticulous’s total raised to $19 million, on top of a $4 million seed round in January 2024.
The company also plans to expand its forward-deployed engineering team, embedding engineers directly with customers to accelerate enterprise adoption and tailor deployments to complex software environments.
As AI agents continue to write larger portions of production software, developers may spend less time writing code and more time deciding whether they can trust it. If that shift continues, the companies verifying AI-generated software could become just as important as the ones generating it.
