01 · Roasts
The CI Boycott
10 repos, 232 tests in geometrical-raytracer alone, 69 in Diffractor — and not a single GitHub Actions workflow file across the entire portfolio. You've built a validation cathedral with no automatic alarm system.
Nine-Minute Masterpiece
cartesian-surfaces-stl-generator: Newton-Raphson solver, surface-of-revolution mesh, STL export, and 5 modules — all committed between 02:35:12 and 02:41:39. Either you pre-wrote it elsewhere or you type at 780 LOC/minute.
Stars: 2. Modules: Infinite.
You have ARCHITECTURE.md, STATUS.md, design.md, and PLAN.md in multiple repos, rigorous BEM reference validation pipelines, and exactly 2 total GitHub stars — both from yourself, presumably.
The License Anarchist
Diffractor, wavec, fractals, cartesian-surfaces-stl-generator — gorgeous optics code released into a legal grey zone. No license means nobody can legally use it, which at 0 forks is technically fine.
Heatmap Archaeologist
40 of 52 heatmap weeks are completely empty, then week 47 onward looks like a physics department deadline hit. Your commit history reads like an ECG with a very late heartbeat.
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% weight60C
- Breadth10% weight65C
- Community10% weight25F
03 · Stats
365-day commit heatmap
65 active days
Language distribution
- Jupyter Notebook43%
- HTML29%
- Lua14%
- C++7%
- Python7%
- TeX1%
04 · Numbers
Owned repos
non-fork
10
Commits
last 12 months
339
Followers
0
Joined GitHub
Jan 2026
05 · Top repos
blancusjh /
geometrical-raytracer
Advanced optical ray-tracing toolkit in Jupyter Notebook format with 12,000 LOC across 9 packages, vectorized numpy propagation, Cartesian stigmatic surfaces, full test suite (232 tests), and design.md + ARCHITECTURE.md documentation. Production-ready research code.
blancusjh /
vecdiff
Specialized Python package for vector diffraction field propagation with structured architecture, comprehensive documentation, and extensive examples. Limited adoption (1 star, no external followers) but scientifically rigorous and well-engineered.
blancusjh /
Diffractor
Typed Python optics library with rigorous validation architecture and geometric exactness, but narrow adoption (1 star, personal project exploring scalar Helmholtz wave diffraction with dual-package monorepo, minimal external footprint).
blancusjh /
wavec
Specialized physics package for vector wave optics on curved interfaces. Well-structured with typed code, comprehensive tests (33 physics tests in tests/), and rigorous mathematical foundation, but nascent (v0.2, 3 commits, created 2 days ago). Clear scientific value for optics research but no external adoption signals
blancusjh /
thesis-about-vectorial-diffraction
Undergraduate thesis repository in Lua/LaTeX with structured chapters, appendices, and figure management. Well-organized with build tooling but limited scope as an academic document project with zero external adoption signals.
blancusjh /
toc-generator
A personal Python tool for extracting table of contents from scanned PDFs using Gemini and PyMuPDF, shipped with minimal tests/CI but typed code and meaningful documentation. Created and pushed within 2 minutes on 2026-06-21, showing experimental single-push pattern.
blancusjh /
cfd-openfoam
Educational OpenFOAM case studies (heat equation, sphere convection, natural convection) with clear structure and README. No tests, CI, license, or typed language. 9MB codebase created 2 days ago with 20 commits.
blancusjh /
voley-dron
Competition submission scaffold for Skydio X2 ping-pong task with instruction.md, data/policy_spec.json, and basic test structure—no license, untyped Python, minimal sustained development (3 days old, 20 commits).
blancusjh /
cartesian-surfaces-stl-generator
Experimental optical lens geometry generator using Cartesian ovoids. Untyped Python with clear README, structured modules, and working demos, but minimal commits and zero external adoption signals in a fresh repo.
blancusjh /
fractals
Educational Mandelbrot fractal renderer comparing three Python approaches (NumPy, Numba, GPU/GLSL). Single-day project with minimal commits, untyped Python, no tests/CI, and acknowledged precision limitations but functional demonstrations.
06 · Timeline
- Jan 15, 2026Joined GitHub
- Mar 13, 2026Created Diffractor
- May 14, 2026Created thesis-about-vectorial-diffraction
- May 25, 2026Created vecdiff
- Jun 21, 2026Created toc-generator
- Jun 23, 2026Created cartesian-surfaces-stl-generator
- Jun 23, 2026Created fractals
- Jul 9, 2026Created geometrical-raytracer
- Jul 15, 2026Created voley-dron — Skydio X2 quadrotor ping-pong policy — MuJoCo, 4 gates + target box
- Jul 18, 2026Created cfd-openfoam
- Aug 8, 2026Created wavec
- Aug 20, 2026Most recent push to Diffractor
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.