01 · Roasts
Jupyter Monolith
64% of your codebase is Jupyter Notebooks and 36% is HTML — that's 100% of your bytes accounted for, leaving Python, Rust, R, and JS rounding to literal 0%. Your language chart is a pie with two enormous slices and five invisible crumbs.
Serial Sprinter
red-flip: built in 2 days. parseplot: 1 commit. monstre: 3 commits, blank README. pyshare: 12 commits, 2 days old. You have 2,220 commits this year but somehow most repos look like they were born on a Friday afternoon and abandoned by Sunday brunch.
Test Desert
10 repos scored. HAS_TESTS=yes: exactly zero. You've got CI pipelines, architectural docs, design.md files, and pre-commit hooks — but not a single test suite committed. The scaffolding for quality is there; the quality itself ghosted.
The License Phantom
Across 52 public repos, licenses are conspicuously absent from the majority of scored projects. You're shipping Rust game servers, Python libraries, and chart parsers into the void with no legal clarity. Open source without a license is just public code that nobody can legally use.
monstre
1 KB. Blank README. 3 commits. Created and last pushed the same day. 'monstre' is the most accurate name for a repo that exists purely to haunt your contributions grid.
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% weight56D
- Consistency20% weight88A
- Quality20% weight57D
- Depth15% weight65C
- Breadth10% weight65C
- Community10% weight65C
03 · Stats
365-day commit heatmap
299 active days
Language distribution
- Jupyter Notebook64%
- HTML36%
- Python0%
- JavaScript0%
- TeX0%
- R0%
04 · Numbers
Owned repos
non-fork
29
Commits
last 12 months
2,220
Followers
203
Joined GitHub
Feb 2021
05 · Top repos
JosephBARBIERDARNAL /
barbierjoseph.com
Personal homepage project with 27 MB codebase and CI/CD pipeline. Documented via design.md, ARCHITECTURE.md, STATUS.md; sustained over ~2 years with 30 recent commits, but minimal external impact (1 star, no forks).
JosephBARBIERDARNAL /
pyshare
Early-stage Python library for reading SHARE epidemiological survey data. Typed, documented with design.md and docs/, CI setup (ruff, type checking), but nascent (0 stars, 1 day old, 12 commits, no tests, no license).
JosephBARBIERDARNAL /
red-flip
Real-time multiplayer Rock-Paper-Scissors game with Elo ranking, actor-based matchmaking, and WebSocket gameplay. Early-stage project (2 days old) with solid Rust architecture but no test coverage, minimal docs, and zero production adoption.
JosephBARBIERDARNAL /
tidytuesday
Personal data visualization portfolio showcasing TidyTuesday entries and custom analyses. Features polars/matplotlib pipelines, reproducible uv setup, and clean visual outputs but lacks tests, CI, license, and architectural documentation.
JosephBARBIERDARNAL /
talk
Archive of personal conference talk materials (slides + videos) spanning Typst, R visualization, and Python packaging. No source code, untyped HTML/assets, 30 commits over ~8 months.
JosephBARBIERDARNAL /
static
37MB codebase with 30 commits over ~20 months, but minimal documentation (README is empty), no tests, no CI, and unknown language. Substantial file size suggests meaningful scope, but lack of typed language and testing infrastructure limits quality assessment.
JosephBARBIERDARNAL /
playing-with-rust
Educational Rust learning exercises with 10 structured lessons covering ownership to lifetimes. Has README, typed Rust code, and organized module structure, but no tests passing, no CI/CD, and minimal original contribution—primarily pedagogical scaffolding.
JosephBARBIERDARNAL /
30DayChartChallenge
A personal 30-day chart challenge portfolio showcasing 5 data visualizations built with Python matplotlib and interactive HTML; minimal scope (9 commits in 3 days), no tests/CI, untyped Python with basic documentation.
JosephBARBIERDARNAL /
parseplot
Experimental Rust/Python matplotlib chart parser using pyo3. Single-commit 683 KB prototype with minimal features, no tests, lacks license, and shows no sustained development history.
JosephBARBIERDARNAL /
monstre
Empty scaffold project created April 4, 2026 with only 1 KB size, 3 commits, blank README titled 'monstre', and no substantive code, tests, CI, or documentation.
06 · Timeline
- Feb 27, 2021Joined GitHub
- Jun 21, 2024Created barbierjoseph.com — My homepage
- Aug 15, 2024Created static
- Mar 4, 2025Created tidytuesday — My contributions to the TidyTuesday challenge, among other visualizations
- Aug 13, 2025Created talk — The material and video of my talks
- Jan 23, 2026Created playing-with-rust — trying to learn rust
- Feb 12, 2026Created red-flip
- Mar 28, 2026Created 30DayChartChallenge — My charts for the 30DayChartChallenge edition 2026
- Apr 4, 2026Created monstre
- Apr 21, 2026Created parseplot — Experimental parsing of matplotlib charts, in Rust
- Apr 23, 2026Created pyshare — Python interface to work with SHARE data (Survey of Health, Ageing and Retirement in Europe)
- Apr 24, 2026Most recent push to pyshare
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.