01 · Roasts
Kernel gravity well
KernelSU has 18,277 stars and 3,996 forks; the rest of the portfolio is mostly trying not to be crushed by its orbit.
C is the house language
92% C says systems work is the main character; TypeScript and Kotlin are allowed cameo appearances.
New-project velocity
freedrop and tango-mirror each landed 5 stars, but both are roughly a day old—ship first, let time do the proving.
Hygiene split-brain
hapi brings tests, E2E, CI, and a license; minime and tiann.github.io arrive without the same paperwork.
Built using
Zoral
Shadows one worker for a week, then takes over their job with zero extra setup. Behaves exactly like the original.
zoral.ai
02 · Category breakdown
- Impact25% weight83A
- Consistency20% weight65C
- Quality20% weight82A
- Depth15% weight65C
- Breadth10% weight65C
- Community10% weight80A
03 · Stats
365-day commit heatmap
217 active days
Language distribution
- C92%
- C++2%
- TypeScript2%
- Assembly1%
- HTML1%
- Kotlin0%
- Other2%
04 · Numbers
Owned repos
non-fork
28
Commits
last 12 months
953
Followers
8,588
Joined GitHub
Apr 2013
05 · Top repos
tiann /
hapi
A widely adopted, typed HAPI monorepo delivering a local-first remote control layer for 10+ coding agents, with web/PWA, Android, CLI, encrypted relay, protocol clients, and substantial automated coverage.
tiann /
KernelSU
KernelSU is a widely adopted Android kernel root platform with a substantial multi-language codebase, structured kernel/userspace/manager components, CI builds, and careful compatibility and security-sensitive implementation.
tiann /
KeyBoxChecker
A focused privacy-first browser tool that locally parses Android keyboxes, verifies certificate chains and key matches, and bundles offline trust data; it has solid automation and tests but limited adoption and no license.
tiann /
freedrop
A shipped, documented P2P CLI/browser file-transfer product with WebRTC, encrypted MQTT rendezvous, resumable transfers, real E2E coverage, and CI, but limited adoption and a very recent history.
tiann /
tango-mirror
A polished, documented npm tool for browser-based Android mirroring, with WebRTC H.264/Opus streaming, encrypted MQTT signaling, shell access, and GitHub Pages deployment; adoption remains modest at 5 stars.
tiann /
homebrew-tap
A focused Homebrew tap with one functional Hapi formula covering macOS and Linux architectures, but minimal documentation, no CI/tests infrastructure, and little visible adoption.
tiann /
minime
A non-trivial Android Pinyin IME with Java UI, Binder decoder service, JNI bridge, and C++ dictionary/search engine, but it has no visible adoption, tests, CI, or repository license.
tiann /
tiann.github.io
A long-lived personal blog repository with 22 stars and 7 forks, but no sampled source files or documented engineering infrastructure; its main strength is apparent maintenance history rather than demonstrated product adoption or code quality.
06 · Timeline
- Apr 23, 2013Joined GitHub
- Sep 21, 2015Created tiann.github.io — my personal blog
- Dec 9, 2022Created KernelSU — A Kernel based root solution for Android
- Dec 24, 2025Created hapi — App for Codex / Claude Code / Pi / OpenCode / Kimi Code / Grok Build, vibe coding anytime, anywhere
- Dec 25, 2025Created homebrew-tap
- Jun 8, 2026Created KeyBoxChecker
- Aug 3, 2026Created minime
- Aug 5, 2026Created tango-mirror
- Aug 6, 2026Created freedrop
- Sep 7, 2026Most recent push to KernelSU
07 · Compare
08 · Rubric
How this score was produced
Overall = Σ (category × weight) + gentle top-end curve
Tier thresholds
▸ How the pipeline works
- 01Scrape.Pull every non-fork repo pushed in the last 90 days, plus your contribution calendar, followers, and language byte counts — straight from GitHub's REST & GraphQL APIs.
- 02Triage.A small model reads every repo's file tree + README and picks the 20 files per repo that actually reveal how you code.
- 03Grade each repo. All repos run in parallel through a fast scoring model that reads the picked files and rates each one independently on Impact, Quality, and Depth — with evidence citations.
- 04Aggregate. A larger reasoning model combines the per-repo scores with server-computed stats (heatmap, commit cadence, language entropy, follower count) to produce the 6-dimension profile score + roasts.
- 05Correct.Deterministic server-side checks enforce anchor-scale floors (e.g. a profile with 2,000+ public commits can't score 30 Consistency) and recompute the final verdict.
~90 seconds per profile, ~$0.25 in compute. Total of ~240 files read across your top-12 repos. One rating per GitHub account per day.
▸ Data sources & caveats
- Heatmap & commit totals: GitHub GraphQL
contributionsCollection— covers the last 365 days, includes private repos when the user has opted in (default). - Language %: byte totals across the top 30 owned non-fork repos.
- Curve: a small upward nudge centered on raw score ≈ 70, capping at 100. Prevents specialists from being unfairly penalised for narrow breadth.
- Anchor corrections: when server-measured signals (e.g. privateWorkLikely, multiRepoVolume, follower count) mandate a minimum category score, the aggregation step enforces it. These are signal-conditional, not identity-based floors.