Your test suite takes 30 minutes. Endform says it can run the whole thing in two. The Stockholm-based startup just raised EUR 1.5 million in seed funding to prove it.

The round was co-led by Alliance VC, Antler, First Fellow Partners, and Greens Ventures, with strategic angels also participating. The March 31 announcement came as the company marks its first year since public launch.

Endform attacks a problem that every engineering team knows but rarely talks about publicly: the CI pipeline bottleneck. As codebases grow and AI coding tools generate more code faster, test suites balloon. Running those tests sequentially means developers sit idle. And idle developers aren't shipping.

The technical approach is elegant in its simplicity. Endform is built around Playwright, Microsoft's open-source browser testing framework. Instead of running end-to-end tests sequentially on one machine, Endform distributes each test to its own cloud machine. Every test runs simultaneously. Your total testing time collapses to the duration of your single slowest test.

Co-founder Jakob Norlin put it in commercial terms: "We decouple the number of tests from the time it takes to run them." That's the key insight. Most CI systems treat test count and test time as linked. Add more tests, wait longer. Endform breaks that relationship entirely.

It sounds obvious once you hear it. Distribute the work, run it in parallel, collect results. But the execution is where it gets hard. Browser-based end-to-end tests are resource-intensive. Each one needs a full browser instance, real rendering, real network calls. Spinning up hundreds of isolated cloud machines, routing tests correctly, aggregating results without flakiness, that's the engineering challenge Endform has spent a year solving.

Among Endform's early adopters is Lovable, the AI-assisted coding platform that's become one of the fastest-growing developer tools in the Nordics. When an AI code generation tool chooses your testing infrastructure, it sends a specific signal: Endform works at the speed that AI-generated code demands.

This is the timing argument for Endform's entire business. AI coding assistants are dramatically increasing the volume of code being written. More code means more tests. More tests means longer CI pipelines. Endform positions itself as the infrastructure that prevents AI-driven productivity gains from being swallowed by AI-driven testing delays.

The investor mix is worth noting. Alliance VC focuses on Nordic B2B software. Antler is one of Europe's most active early-stage investors. First Fellow Partners and Greens Ventures round out a syndicate that covers different stages and networks. For a seed round, four co-leads is unusual. It suggests competitive demand.

Stockholm has quietly become a serious hub for developer infrastructure. Vercel has a significant Nordic engineering presence. Several developer tools that became category leaders, Spotify's open-source contributions, the entire Backstage ecosystem, emerged from Swedish engineering culture. Endform fits this lineage.

The broader market for testing automation is estimated at several billion dollars annually, but most of that spending goes to legacy tools. Endform's bet is that the AI code generation wave creates a new category of testing infrastructure, one designed for volume and speed rather than just coverage and compliance.

Whether EUR 1.5 million is enough to build that category is the open question. Seed rounds in developer tools often need to demonstrate compelling unit economics before Series A. If Endform can show that customers reduce CI pipeline time by 90 percent and stick around, the next round writes itself. If the parallel execution model proves harder to scale than expected, or if Playwright's market share shifts, the path gets harder.

The pitch isn't really about testing. It's about developer velocity. Every minute a developer waits for tests is a minute they're not building. In a world where AI tools generate code in seconds, the testing pipeline becomes the new bottleneck. Endform bets that bottleneck is worth EUR 1.5 million to solve. Given the trajectory of AI-assisted development, they're probably right. The question is just how big the prize gets.