01 · Roasts
222 PRs, All Self-Inflicted
totalPRsYear = 222 with soloPct = 100%. That's not contribution activity, that's you reviewing your own homework and calling it a pull request. Open one PR on someone else's repo and the number actually means something.
Python Monoculture
93% Python. Your langPcts look like a pie chart that forgot it was a pie chart. Java clocks in at 2% — basically a rounding error from one capstone. Broaden or commit to the snake.
0 Followers, 0 Following, 0 Stars
Two stars total, zero followers, zero following. You're building in a sealed bunker. OrientAI is live on Railway and submitted to an INRIA challenge — tell literally anyone about it.
OrientAI_2: Two Commits, 2.5 Minutes
An entire repo created and finalized in 150 seconds. That's not a project, that's a file you accidentally put in the wrong folder and gave a README to out of guilt.
Burst-and-Ghost Pattern
Weeks 31–40 show dense 3–4 activity days, then weeks of silence. The heatmap reads like a student who discovered GitHub deadlines exist. The OrientAI sprint was real — keep that energy year-round.
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% weight60C
- Consistency20% weight65C
- Quality20% weight72B
- Depth15% weight65C
- Breadth10% weight55D
- Community10% weight25F
03 · Stats
365-day commit heatmap
80 active days
Language distribution
- Python93%
- Java2%
- TypeScript1%
- JavaScript1%
- C1%
- CSS0%
- Other2%
04 · Numbers
Owned repos
non-fork
16
Commits
last 12 months
768
Followers
0
Joined GitHub
May 2022
05 · Top repos
matjussu /
OrientAI
French educational RAG system (47k+ records) with production discipline: FastAPI, FAISS index, Mistral LLM, 2500+ pytest tests, comprehensive docs (ARCHITECTURE.md, DECISION_LOG.md), CI/CD, and deployed platform at orientai-platform.fr.
matjussu /
Picsou_back
Spring Boot 3.5 personal finance REST API (M1 MIAGE final project) with JWT auth, multi-account transactions, roommate split tracking, WebSocket coloc events, AI insights (Claude), and comprehensive PostgreSQL schema—deployed live on Render with typed code, structured 260KB codebase, and integration tests using Testcon
matjussu /
MSU
Pedagogical site documenting four coordinated Claude agents; small-scope project with meticulous safety & validation patterns, no external adoption.
matjussu /
SmartTraveler
TypeScript Next.js travel optimization mockup with structured design system, documented constraints, and multi-route app. Typed, documented, ~7KB codebase with no tests/CI—solid indie portfolio project but no production adoption.
matjussu /
Site-les-100
Custom artisanal bakery website built with vanilla HTML/CSS/JS and GSAP animations; 92MB repo with structured product data, PWA support, and a working e-commerce flow, but no tests/CI and targeting a niche local business with zero adoption signals.
matjussu /
Picsou-front
Polished M1 university final project: budget app for 17–26 year olds with Angular 19, real-time coloc expenses via STOMP, AI insights. Typed, documented, structured. Deployed to Vercel; 22 of last 30 commits.
matjussu /
Site-ReadKode
Professional landing page for ReadKode mobile app built with Astro 5, React 19, TypeScript, and Tailwind CSS. Fully typed, bilingual (FR/EN), CI/CD configured, but zero stars/forks and no tests. Emerging indie project with clean architecture and documented structure.
matjussu /
OrientAI_2
Design reference pack for OrientAI frontend platform; handoff guide for Claude Code with design system specs, no functional codebase, created and pushed same day (21 KB, 2 commits in 30min window).
06 · Timeline
- May 29, 2022Joined GitHub
- Apr 21, 2025Created Site-les-100
- Dec 27, 2025Created Site-ReadKode — Site vitrine de l'app ReadKode
- Apr 10, 2026Created OrientAI
- Apr 19, 2026Created Picsou_back
- Apr 19, 2026Created Picsou-front
- May 20, 2026Created SmartTraveler
- Jun 9, 2026Created OrientAI_2
- Aug 9, 2026Created MSU — Site pedagogique : comment fonctionne un setup de quatre agents Claude Code coordonnes
- Aug 21, 2026Most recent push to Site-les-100
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.