Blog
The builder's perspective on BYOK AI economics, modern QA tooling, MCP for testing, and the gap between dev tools and QA workflow. Honest, opinionated, written by someone shipping in this space.
Coming soon
First posts drop alongside our public alpha announcement. Want them in your inbox? Apply for the design partner program and we'll send drafts early.
The story of how a frustrated QA architect ended up building yet another test management tool — and why "yet another" is wrong this time.
Math on what your team actually pays for AI test generation under credit-metered TMS pricing vs BYOK. With real numbers, not marketing.
How an MCP server can scan your codebase, generate a Gherkin plan, sync to your TMS, and run Playwright — all from a single Claude Code chat.
Why every TMS shipping a separate npm package per framework is solving the wrong problem — and how a single uploader changes the math.
How LLMs choose between Gherkin and step-by-step format, what ISTQB rules to feed them, and the prompts that actually produce useful output.
A practical guide for compliance-heavy teams that can't ship test data to a SaaS TMS — and why Coolify makes this 10× easier than it used to be.
Why pass-rate alone lies, how to compute a real flaky ratio across CI runs, and the threshold that separates a healthy suite from a fragile one.
The pattern: poll Linear / Jira for status changes, auto-update test results, fire a notification when retest is due. The plumbing matters.
When each category wins, when each loses, and how to pick based on your team's actual workflow — not the marketing pages. Honest, opinionated, and updated as the space evolves.
Read the drafts first
Alpha is free for everyone. No credit card. Sign up takes 30 seconds.