Endform raises €1.5M to speed up tests and relieve CI in the age of AI code

Swedish startup Endform has raised €1.5M in seed funding to take on one of the most irritating problems in modern software development: waiting on CI pipelines. Amid the boom in AI code generation, that delay is becoming more visible. Teams can write and change code faster and faster, but verification increasingly fails to keep pace.

The CI bottleneck

Endform's idea rests on a simple reality: for a long time, what has slowed development down is not writing code but the time between a commit and confirmation that everything has passed the tests. When a company speeds up feature delivery, the volume of automated checks grows too. Add AI tools that help developers produce more changes in the same amount of time, and the CI queue quickly becomes a constant source of downtime. The team has seemingly done its work, but the release effectively hangs on waiting for a green status.

The problem is not new, but it is getting costlier. Where a slow pipeline once merely irritated, it now directly eats into the gains from AI coding. Code generation shortens the time to a pull request, but it does not remove the risk of breakages, regressions and flaky tests. The bottleneck simply shifts: instead of a shortage of engineering time, companies face a shortage of validation speed. Endform is trying to step into exactly that gap.

"Code review has always had a bottleneck — not the code itself, but

the waiting."

A bet on testing

Endform is a Swedish startup in the testing segment, and the €1.5M it raised is meant to help it attack the bottleneck at the level of CI processes. The published description suggests the company is betting not on yet another code generator but on the infrastructure layer around it: on what happens after changes are written and before they reach production. This is a less visible part of development, but it is where time gets lost, especially in teams with many automated tests.

Looking at the pain itself, teams expect very specific results from such solutions:

• cut test run time without losing coverage
• prioritize the most important checks for each change
• find flaky and duplicate tests faster
• reduce the number of reruns and idle waiting
• return feedback to the developer closer to the moment the code was written

That is why even a relatively small seed round looks timely. In infrastructure products, value often comes not from a loud storefront but from steadily eliminating delays that every team feels daily. If Endform can prove it reduces waiting time without compromising quality, it will have a clear argument for CTOs and engineering managers: the same engineering staff starts delivering results faster.

Why now

By infrastructure-market standards the round is not huge, but for an early-stage company it makes sense: the problem is clear, the market is only getting more acute, and the benefit is easy to count in hours of waiting, engineer costs and release speed. The more actively companies adopt AI assistants in the IDE and code generation, the sharper the contrast between fast code creation and a slow verification loop.

For teams, this is no longer abstract optimization but a matter of economics. If every pull request sits in the queue longer, context switching increases, reviews are delayed, and developers start splitting changes up or, conversely, piling them into larger batches. Both scenarios make releases less predictable. That is why tools that speed up the testing and CI layer have a chance to become a required part of AI-first development rather than a niche DevOps add-on.

What this means

Endform is trying to earn money not from the hype around AI code itself but from its side effect: an overloaded system for verifying changes. If the startup can really cut CI waiting without losing quality, the winners will be not only testing teams but the entire development cycle from PR to release.