🏆 Notable contributions
Popular open-source repos they've shipped to (commits + PRs)
Score breakdown
🛠 Featured work
Their own popular and pinned repositories
🧬 Stack & domains
🧬 Most similar developers
Closest profile, nearby score
🔥 Full roast
🔥 13 original repos net just 13 stars total, half your popular-repo contributions are docs/tutorials, and 15 external PRs were self-closed—your GitHub resume is fully packed, but your actual code substa
PopSoda2002 — 82.00/100 · ELITE (Elite · Top-tier dev)
TL;DR: Prolific contributor to top ML infra repos, but your star-starved home turf and docs-heavy contribution mix reveal you’re a polished open-source tourist, not a homegrown builder.
| Dimension | Score | Notes |
|---|---|---|
| Account maturity | 9.7/10 | Registered 5.71 years ago, with contributions spread across 7 calendar years — you’ve been consistent long enough to stop counting active months like a new grad. |
| Original project quality | 8.6/18 | 13 original repos net just 13 total stars, with your top project PopSoda2002/FSDP_memory_estimator earning a 0.9 quality score for its functional demo but only 7 public stars. |
| Contribution quality | 23.3/27 | 57 merged PRs out of 80 total, with a low 8% maintainer rejection rate; 5 maintainer-closed unmerged PRs, 15 self-closed external PRs, 0 self-closed own-repo PRs. |
| Ecosystem / maintenance impact | 20/20 | All-time 70 PRs + 114 commits across 9 popular repos (including 1 PR to llama.cpp, 8 PRs + 7 commits to sglang, 8 PRs + 6 commits to verl); the 12 verified PRs are a file-level quality sample, not the full contribution count. |
| Community influence | 4.2/8 | 64 followers, following 84 — you’re still in the simp league of open source, following more people than inspire others to follow you. |
| Activity authenticity | 16.2/17 | 977 contributions last year, 0 days since last activity, no PR flooding, templated PR, or star inflation flags. |
Red flags
- Nearly 60% of your all-time popular-repo contributions target tutorial and documentation repositories (63 commits and 15 PRs to zhaochenyang20/Awesome-ML-SYS-Tutorial alone), with a recent external doc-like PR ratio of 0.54, just under the 0.55 threshold that would cap your impact score — most of your visible external work is polishing docs and tutorials rather than core engineering.
- Your original projects have almost no public traction: your 13 original repos have a combined 13 stars, with your highest-rated project (FSDP_memory_estimator) sitting at just 7 stars, so your home turf offers no proof you can build something people want to use.
- 15 of your 20 closed unmerged external PRs were self-closed by you, suggesting a meaningful share of your contributions were either superseded or did not align with maintainer requirements before you pulled them yourself.
Score calibration No extra adjustment. The base score already accounts for your docs-heavy contribution mix, low original project traction, and self-closed PR pattern.
Verdict Normal high-volume open-source contributor. You’re a reliable gig worker for top ML projects, but your own repo graveyard means you’re still building someone else’s resume, not your own.