01 · Roasts
Speedrun Architect
marginal_edge shipped its entire 5-stage pipeline, bootstrap CI engine, and vitest suite in a 12-minute commit window. quant built 265 MB of trading infrastructure in under a day. These aren't projects — they're AI-assisted lore drops.
CI? Never Heard of Her
Zero CI pipelines across all 11 repos. Not one. You've written Needleman–Wunsch alignment algorithms and multi-stage transformer training loops, but 'npm test on push' remains an unsolved problem.
Graveyard Groundskeeper
51% of your repos haven't been touched in 2+ years. You have 56 public repos and 22 followers — that's a ratio that suggests you're building for the void.
The Prolific Abandoner
Thea (voice desktop app), Yapper-Voice (also voice?), replay (screen recording), breeze-desktop-releases (empty bucket) — you have at least three overlapping abandoned desktop-app attempts and a releases repo with nothing in it.
Solo to a Fault
97% solo commits, 3 external PRs all year, following exactly 2 people. You've built a quant trading system, an LLM interpretability decoder, and a GitHub rating engine — but apparently no one else exists on this platform.
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% weight62C
- Consistency20% weight65C
- Quality20% weight72B
- Depth15% weight58D
- Breadth10% weight65C
- Community10% weight55D
03 · Stats
365-day commit heatmap
132 active days
Language distribution
- TypeScript77%
- Python9%
- JavaScript7%
- TeX2%
- HTML2%
- CSS1%
- Other2%
04 · Numbers
Owned repos
non-fork
51
Commits
last 12 months
195
Followers
22
Joined GitHub
May 2020
05 · Top repos
AryaaSk /
ratemygithub
TypeScript arcade-style GitHub profile scoring system with three-pass AI pipeline, Zod validation, Drizzle ORM, structured API and comprehensive rubric implementation. Non-trivial shipped product with typed code and documentation.
AryaaSk /
residual_stream_visual_decoder
Research project on visual decoding of LLM residual streams: trains Gemma 4 E2B to produce stroke drawings from activations. Well-documented multi-stage training pipeline with working artifacts. Early-stage research, ~5.5 days old, limited adoption but substantial technical depth.
AryaaSk /
quant
Multi-market transformer trading POC with 265MB codebase, agent-scraped text embeddings, and honest post-mortem (LIMITATIONS.md) revealing constant-predictor collapse. Infrastructure-complete but single-feature mean-reversion driver; experimental methodology.
AryaaSk /
cs_tripos_partia
Personal adaptive exam prep system for Cambridge CS Tripos Part IA with AI-powered question generation and marking. 87 MB codebase with typed Python, structured architecture, and comprehensive documentation (README + design.md + ARCHITECTURE.md), but no tests, CI, or license—active portfolio work from owner with multip
AryaaSk /
replay
Pre-release macOS screen recording → AI bug report tool. TypeScript + Rust desktop app with structured docs, typed sidecar pipeline, and documented architecture. Early-stage hobby project, not yet shipped widely.
AryaaSk /
Actual_WPM
Novel typing test app using LLM-assisted scoring to measure real-world typing speed. Early-stage personal project with typed TypeScript, structured Next.js layout, and clear README explaining the three metrics (raw/true/actual WPM), but minimal ecosystem signal and only 4 recent commits over a single day.
AryaaSk /
Thea
TypeScript Electron desktop app for voice-driven automation via OpenClaw gateway. Well-typed, modular architecture with WebSocket client, IPC handlers, and React UI. No tests, CI, or docs. Fresh repo (1 day old) with 25 commits.
AryaaSk /
marginal_edge
Educational betting prediction testbed with full 5-stage pipeline (data→model→decision→simulation), shipped with worked Liverpool example, rigorous statistical metrics (log loss, Brier, CLV with bootstrap CIs), comprehensive docs, and tests.
AryaaSk /
Zootropolis
TypeScript AI company simulation project with 30 commits over 3 days, featuring tests and comprehensive documentation (docs/, design.md, ARCHITECTURE.md, STATUS.md) despite no README, but minimal adoption signals.
AryaaSk /
Yapper-Voice
Newly created TypeScript project (3 commits in ~2 hours) with design documentation but no source code samples, tests, CI, or license. Early-stage experiment with aspirational docs but minimal shipped code.
AryaaSk /
breeze-desktop-releases
Empty release artifact repository with minimal README. Zero stars, no files, single commit in sampling. Appears to be a release dump for an Electron app with no independent value.
06 · Timeline
- May 3, 2020Joined GitHub
- Jan 25, 2026Created breeze-desktop-releases — Desktop releases from Electron Builder for Breeze
- Mar 7, 2026Created Thea
- Mar 12, 2026Created Yapper-Voice
- Apr 2, 2026Created cs_tripos_partia — Agent to help prepare for CS Tripos Part IA exams
- Apr 11, 2026Created Actual_WPM — stop fixing your typos. the llms already know what you meant
- Apr 15, 2026Created Zootropolis — Full AI run companies
- Apr 18, 2026Created ratemygithub — Rate your GitHub against others!
- Apr 27, 2026Created replay — Click record. Show your bug. Get a perfect description for your AI agent.
- May 14, 2026Created marginal_edge — Educational testbed: can your model beat a betting market? Build a pipeline, train a predictor, simulate the bets.
- May 16, 2026Created quant — Multi-market transformer trading POC with agent-scraped text and Voyage embeddings
- May 19, 2026Created residual_stream_visual_decoder — Visual lens into LLM residual streams: stroke-output decoder of Gemma 4 E2B activations, NLA-style autoencoder with vision-pathway reconstruction. Research project.
- May 22, 2026Most recent push to residual_stream_visual_decoder
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.