Automated, hosted UI tests with AI test discovery (octomind.dev)
I am Marc, the co-founder of octomind.dev. I am a former chess player turned machine learner turned entrepreneur. I always wanted to develop products based on the latest AI research. I did it at my previous startup with computer vision and now at octomind.dev with LLMs/agents. Octomind is an AI-powered end-to-end testing platform for web apps. You enter your URL, and we will auto-generate, run, and auto-maintain UI tests. My co-founder Daniel and I used to work together in our previous AI startup. We had never-ending problems with end-to-end testing. We tried many frameworks and tools…
In plain words
Octomind is an AI-powered end-to-end testing platform for web applications. Users provide a URL, and the platform automatically generates, executes, and maintains UI tests without manual coding. Designed for development teams struggling with traditional testing frameworks, it uses large language models and AI agents to reduce the burden of writing and maintaining tests while keeping costs reasonable.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
I am Marc, the co-founder of octomind.dev. I am a former chess player turned machine learner turned entrepreneur. I always wanted to develop products based on the latest AI research. I did it at my previous startup with computer vision and now at octomind.dev with LLMs/agents. Octomind is an AI-powered end-to-end testing platform for web apps. You enter your URL, and we will auto-generate, run, and auto-maintain UI tests. My co-founder Daniel and I used to work together in our previous AI startup. We had never-ending problems with end-to-end testing. We tried many frameworks and tools and even looked into outsourced QA services. Nothing that wasn’t a pain for our dev teams to write and maintain wouldn’t hold back releases or be reasonably priced. We have a decent level of experience in baking AI into software products. It made sense to apply it to e2e testing. Whenever a test target is set, our AI agents discover user flows and interactions leading toward a desired user goal. This serves as the foundation for creating deterministic test cases. We use it for automated test case discovery and are working on expanding it to maintenance so that broken tests can self-heal. We’ve built Octomind on top of Playwright, which is imho the most advanced framework for end-to-end testing today. We don’t need access to your code, and the generated Playwright code is yours. We run and host the tests. We report the test results in our app or directly in PR comments if you integrate Octomind with your CI pipeline. To help with debugging, we’ve integrated Trace Viewer and created Debugtopus - an open-source tool for local debugging. (see https://github.com/OctoMind-dev). We’re still in an early access mode, so any feedback from the community would be appreciated. Try us out for free. We want to keep the freemium option even as we grow.
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
