Gabriel Spencer-Harper likes to tell a story that makes software engineers wince. A perfectly functioning feature ships to production. Users click around. Everything seems fine—until someone notices a button has vanished, or a checkout flow silently breaks on Safari, or a modal window renders as a blank rectangle on mobile.
By then, the damage is done.
It's the kind of bug that slips through even rigorous code review, particularly now that AI is churning out pull requests faster than human eyes can parse them. And it's why Meticulous, the London startup Spencer-Harper founded with his brother Quentin, just closed a $15 million Series A round. The announcement came July 15, 2026, led by Chemistry with Menlo Ventures joining in—alongside an angel roster that includes Lachy Groom, Jason Warner, Arash Ferdowsi, Guillermo Rauch, Calvin French-Owen, and Jason Ginsberg.
Lachy Groom, once of Stripe. Jason Warner, now co-founding Poolside after running engineering at GitHub. Arash Ferdowsi from Dropbox's early days. Vercel's Guillermo Rauch. Calvin French-Owen, who helped build Segment before a stint at OpenAI. Jason Ginsberg, formerly Cursor's head of engineering, now at xAI. The list goes on.
That kind of investor lineup doesn't materialize by accident. It suggests something deeper than a clever product—perhaps a problem these engineers have all felt personally.
When Code Moves Faster Than Testing Can Keep Up
Meticulous tackles what the founders call "the last bottleneck in software velocity." The tool automates end-to-end frontend testing by recording actual user sessions—thousands of them—then replaying those interactions deterministically to catch visual regressions before anything ships. Think of it as a digital quality assurance team that never sleeps, never gets bored, and scales infinitely.
The pitch is straightforward, almost blunt: AI writes code now. Fast. Humans review what they can. But testing? Testing lags behind, creating a dangerous gap where bugs breed.
Gabriel, who was a software engineer at Dropbox from 2019 to 2020 and at Opendoor from 2018 to 2019, saw this firsthand. Quentin logged roughly a decade at Palantir leading frontend work. The brothers emerged from Y Combinator's Summer 2021 cohort with a thesis that felt both obvious and ambitious—automate the tedious, error-prone work of verifying that user interfaces actually work.
By July 2026, the company claims to be processing over 90,000 pull requests monthly. That translates to 1.5 billion snapshots—a scale that suggests either rapid adoption or deeply embedded usage. Likely both.
The Customer Roster Tells Its Own Story
Notion uses it. So does ElevenLabs, the voice AI startup. Dropbox, where Gabriel once worked. Wiz, the cloud security unicorn. LaunchDarkly. Mercor. CoreWeave, the infrastructure player that's become shorthand for AI compute scale.
"We piloted the tool, were immediately impressed and rolled it out across the entire engineering organization," said Notion's Head of Developer Experience in a company statement. The tool has become what they call an "essential guardrail."
A director at LaunchDarkly went further, describing it as "the best regression tool you could imagine"—one that lets teams "move fast and raise the quality bar" simultaneously, a combination that usually feels like wishful thinking in software development.
It's the kind of customer traction that makes investors pay attention. Not just users kicking the tires, but engineering teams at well-respected companies weaving the product into their daily workflow.
What Comes After Frontend
The fresh capital will fund what you'd expect: more engineers, deeper product development. But according to coverage from Vestbee and Tech Funding News published around the funding announcement in mid-July 2026, Meticulous has its sights set beyond frontend testing. Backend validation is next. Full-stack testing after that.
It's an ambitious roadmap, though perhaps not surprising. Once you solve one piece of the testing puzzle, the rest starts looking solvable too.
Meticulous previously raised $4 million in seed funding—that round was announced in January 2024, led by Coatue with Y Combinator, Base Case Capital, and Soma Capital participating. The company's LinkedIn profile suggests a team somewhere between 11 and 50 people, a range that's deliberately vague but typical for startups in this phase.
A Broader Wave, or Just Noise?
The timing is worth noting. Meticulous isn't alone in betting that testing infrastructure is ripe for reinvention. Momentic, another player in the space, raised its own $15 million Series A back in November 2025, that one led by Standard Capital. Autosana secured $3.2 million in February 2026 for what it calls "agentic QA" across mobile and web platforms.
Three funding announcements in less than a year, all circling the same problem from slightly different angles. Either venture capitalists are chasing a trend, or testing really has become the constraint everyone underestimated.
Given the caliber of angels backing Meticulous—engineers who've built and scaled some of the most complex software systems in the world—it's probably the latter. These aren't investors who get excited about buzzwords. They're operators who've felt the pain.
Which means the Spencer-Harper brothers might be onto something more than just a funding moment. They might have found the bottleneck that no amount of AI-generated code can bypass.
At least not yet.
