Research Paper 24 pages · PDF

Predicting Defects
Before They Happen

A validation study of Calyntro's Git-history risk signals across three open-source projects: MongoDB (C++), uv (Rust), and Qt Framework (C++). Can historical churn, complexity, and ownership data reliably predict where the next bugs will appear — before the code is written?

Qt Framework · C++
61%
of future bug fixes in the top-risk 26% of files (AUC 0.792)
uv · Rust
40%
of future bug fixes in the top-risk 25% of files (AUC 0.730)
MongoDB · C++
42%
of future bug fixes in the top-risk 29% of files (AUC 0.619)

Contents

  • 1The Problem — Why code review and coverage metrics miss defect concentration
  • 2Methodology — Churn, Complexity, Hotspot Score and the prediction window
  • 3The Ownership Paradox — Concentration vs. Silo Risk as orthogonal dimensions
  • 4Results — AUC comparison, top-quartile concentration, signal ranking by project
  • 5Implications for Engineering Leaders — When to trust each signal
  • AAppendices — Full project breakdowns, Repowise benchmark comparison, history-length dependency

Download the full whitepaper

We'll send the PDF link directly — no newsletter, no follow-up unless you want one.

No spam. One email max.
✓

Thanks — here's the whitepaper.

You can also bookmark this link directly for future reference.

Download PDF (24 pages)

Calyntro · Predicting Defects Before They Happen · 2026