One change lights up the city. We run only the tests that visit it.
Every function is a light in the city. When one changes, we follow code coverage and requirement traceability to the exact test cases it affects, score them by risk and cost, and run the highest-value set within your time budget.
🌆 The living city
Your product is thousands of functions — the smallest units of code, each doing one job. Coverage records which tests exercise each one.
✨ A change lands
A commit modifies a handful of functions. We diff it against the previous revision to pinpoint exactly which functions changed.
🔌 Trace the coverage
The coverage map already links every function to the tests that execute it. We pull in each test case that covers a changed function.
📋 Trace the requirements
When a change touches a requirement, we follow its traceability links and add every test case that verifies it.
🧠 Recall the history
For each candidate we load its execution history: typical run time, and how recent runs passed or failed.
📊 Score every candidate
Each candidate is scored from four signals — relevance to the change, run time, recent failure rate, and time since last run — combined into one number.
⏳ Fit the time budget
You set a time budget. We add test cases from the top of the ranking until the budget is full — highest value first.
⏱️ Run only what matters
Instead of the full 71-test regression suite (~38 minutes), you run a focused subset targeting the change — most of the confidence in a fraction of the time.