Score breakdown
🧬 Most similar developers
Closest profile, nearby score
🔥 Full roast
🔥 Self-proclaimed depth estimation researcher? Your 28 original repos have a combined 22 stars, and you've modified big corp's CI more times than your own projects have stars total — nice GitHub resume
Haruko386 — 75.40/100 · SOLID (Solid · Trustworthy)
TL;DR: Solid backend contributor for InfiniFlow's 83k-star RAGFlow who ships real code, but his own project portfolio is a starved wasteland of learning exercises and 1-star throwaways — great employee open-source profile, weak independent one.
| Dimension | Score | Notes |
|---|---|---|
| Account maturity | 6.4/10 | Registered ~3 years ago with 4 years of logged contribution activity; consistent presence but no standout tenure milestones. |
| Original project quality | 8.7/18 | 28 original repos with a combined 22 total stars; top repo ApDepth is a legitimate depth estimation research codebase with 12 stars, but the rest are mostly learning projects and 1-star novelties. |
| Contribution quality | 26.1/27 | 119 merged PRs out of 146 total (2% maintainer rejection rate); 16 self-closed external PRs are just abandoned drafts, not rejections. 84 of 86 verified impact PRs are core backend code (handlers, services, routers, tests) with only 2 docs/site-related PRs in the sample. |
| Ecosystem / maintenance impact | 14.8/20 | All-time 107 PRs + 94 commits into 3 popular repos totaling ~86k stars, 98% of volume hitting InfiniFlow/ragflow; work is substantive backend engineering, not docs/CSS fluff, but contribution diversity is almost zero. |
| Community influence | 3/8 | 9 followers, 15 following — a following ratio that screams "I'm here to code, not to collect an audience". |
| Activity authenticity | 16.4/17 | 1132 contributions last year, last active today, no signs of botting or burst farming; activity is genuine, even if heavily slanted toward employer work. |
Red flags
- Bulk low-effort contributions pad your PR count: multiple 5k-15k line PRs that only update
conf/all_models.jsonwith new model entries, adding zero meaningful engineering depth. - 70% of all PRs target your employer's InfiniFlow/ragflow repo — while perfectly legitimate, this turns your GitHub profile into a company open-source gig worker resume, not an independent community contribution record.
- You've submitted 4+ iterative PRs to the same
.github/workflows/release.ymlfile, suggesting you're firefighting platform-specific CI bugs ad-hoc instead of designing a robust, cross-platform CI pipeline.
Score calibration No extra adjustment. The base score already prices in your genuine core engineering work on RAGFlow, legitimate ApDepth research code, near-zero PR rejection rate, and authentic activity. The minor caveats (single-repo concentration, low-complexity bulk JSON PRs, repetitive CI fixes) don't cross the threshold for a score bump or haircut.
Verdict Normal, solid contributor. You ship real, merged backend code for a major open-source project and have legitimate original research work to your name, but your profile reads like a company open-source gig worker with a side of learning-project clutter — not a bad look, just not a standout independent open-source profile.