$15m · Series A · Software · London, UK
Meticulous has raised a $15m Series A led by Chemistry, with participation from Menlo Ventures and investors including Lachy Groom, Jason Warner, Arash Ferdowsi, Scott Belsky, Guillermo Rauch, Calvin French-Owen, Caitlin Colgrove and Jason Ginsberg. The company announcement says the financing will support its effort to make software testing exhaustive enough for engineering teams to ship at the pace of AI-generated code.
Turning real sessions into deterministic tests
Meticulous records how developers and users interact with a web application, then turns those sessions into an automatically maintained suite of end-to-end flows. When a pull request arrives, the platform selects sessions that cover the affected features, replays them in parallel browsers and compares visual snapshots against the base version. It is designed to catch behavioural, logical and visual changes without developers writing individual assertions.
The company’s key engineering choice is a deterministic browser intended to remove timing and rendering noise. Meticulous says some changes can be replayed across as many as 10,000 browsers, while algorithms curate overlapping sessions and warn when edited code is not covered. That makes the product’s commercial value depend on coverage and signal quality: teams pay to reduce the time spent authoring tests and investigating flaky failures, but only if the reported differences are trustworthy enough to block a release.
AI code makes review the bottleneck
Meticulous names Dropbox, Notion, Wiz and ElevenLabs among customers using the platform across large codebases. Notion says it rolled the tool out across its engineering organisation after a pilot, while LaunchDarkly frames code review and feedback as the constraint emerging as code generation becomes cheaper. Those are customer and company claims, but they show where Meticulous sits in the development workflow: after code is produced and before a human approves the change.
The Series A funds a specific scaling test. Meticulous must expand the number and complexity of flows it can replay without making test runs too slow or expensive, maintain deterministic behaviour as browsers and applications change, and help customers handle recorded-session data safely. If it can preserve confidence at that scale, testing becomes infrastructure for agentic development rather than a manual step that grows with every new feature.
