Issue Triage Bot for a Popular Eclipse Foundation Project
Overview
What this challenge is about.
Read the project's existing labels, issue templates, and last 800 issues. Build a Go (or Python) GitHub Action that runs on issue_opened and issue_edited events. Implement heuristics: component label from a regex/keyword map, severity from explicit keywords + stack trace presence, reproducibility from template-field completeness, duplicate-likelihood via TF-IDF (Term Frequency-Inverse Document Frequency) similarity against the last 1,000 issues. If repro steps are missing, post a templated comment requesting them and add a needs-info label. Run a 4-week pilot on 160 incoming issues. Deliver: bot source repo (PR-ready), 8-page heuristics document, 6-page 4-week pilot evaluation with precision/recall on the labels.
The Brief
What you'll do, and what you'll demonstrate.
Build a deterministic, LLM-free issue-triage bot for an Eclipse Foundation project that survives a 4-week pilot on 160 issues.
Earning criteria — what you'll demonstrate
- Build a GitHub Action with non-trivial heuristic logic
- Implement TF-IDF duplicate detection without an ML dependency
- Author templated maintainer comments that get repro info without sounding rude
- Evaluate triage precision/recall against maintainer ground truth
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
Career mappings coming soon.