01 · Roasts
Notebook Maximalist
86% of your codebase is Jupyter Notebooks. You're essentially a very sophisticated .ipynb file with a GitHub account attached.
Sprint God, Endurance Zero
Arcane-PP: born and died in 8 hours. Arcane-OCR: 1 day. Hailo-Training: 24 hours. Your commit history reads like a series of Red Bull-fueled all-nighters that never got a morning-after.
The Arcane Cinematic Universe
You have Arcane, Arcane-PP, Arcane-OCR, and Arcane-GLM. That's four repos sharing a name and a collective total of 0 stars. The franchise is not yet a blockbuster.
58 PRs, 4 Followers
You filed 58 pull requests this year but only 4 people follow you. Either you're exclusively PRing yourself, or you're the most productive ghost on GitHub.
Architecture Doc Collector
README.md, design.md, ARCHITECTURE.md, STATUS.md — you write more docs about the system than you write the system. Information-Lab has more documentation tiers than commits per week.
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% weight60C
- Quality20% weight72B
- Depth15% weight65C
- Breadth10% weight55D
- Community10% weight40D
03 · Stats
365-day commit heatmap
128 active days
Language distribution
- Jupyter Notebook86%
- HTML12%
- Rust1%
- Python1%
- TypeScript0%
- Shell0%
04 · Numbers
Owned repos
non-fork
12
Commits
last 12 months
213
Followers
4
Joined GitHub
Oct 2022
05 · Top repos
Rah-Rah-Mitra /
Information-Lab
Edge-native multi-agent knowledge-graph pipeline converting PDFs to Obsidian vaults with concurrent research lanes (curator, bridge, theorem, derivation). Typed Rust + SQLite + structured schemas, full CI/CD, 30 commits in 7 days shows focused burst work.
Rah-Rah-Mitra /
Arcane
Arcane is a typed Rust local-first research archival app with structured architecture (src/ subdirs, 450 KB codebase), CI/CD, complete docs (README + design.md + ARCHITECTURE.md), and sophisticated PDF handling (outline recovery, clustering, offset detection) but lacks tests and sees personal-project adoption levels.
Rah-Rah-Mitra /
Hailo-Training
Personal research/training project using YOLO26n for document layout detection on Hailo edge hardware. Includes 5 Jupyter notebooks spanning dataset prep, ONNX export, quantization, compilation, and benchmarking; structured well with typed Python, comprehensive docstrings, and docker setup.
Rah-Rah-Mitra /
Arcane-OCR
Hardware-accelerated OCR pipeline for Raspberry Pi + Hailo-8L NPU with PaddleOCR models, tiling strategies, edge fusion, and SymSpell correction. Structured Python codebase with comprehensive design documentation but no tests, CI, or license.
Rah-Rah-Mitra /
Portfolio
Personal portfolio website built with TypeScript/React/Vite showcasing AI/ML and security achievements. Features dual-theme UI with physics interactions, theme switching, and project showcases, but no tests or CI. Typed, documented, structured codebase.
Rah-Rah-Mitra /
Arcane-GLM
Young Python project (2 days old, 2 commits) building a 6-stage document OCR pipeline for hierarchical TOC extraction. Typed, well-documented (README + ARCHITECTURE.md + design.md), structured codebase with tests, but no CI and no license yet. Experimental stage.
Rah-Rah-Mitra /
Arcane-PP
Personal PaddleOCR wrapper with deterministic TOC parsing for document extraction. Typed Python, structured layout, comprehensive tests, but minimal adoption (0 stars, unfinished code, created today).
06 · Timeline
- Oct 10, 2022Joined GitHub
- Jul 1, 2025Created Portfolio
- Mar 14, 2026Created Arcane
- Mar 19, 2026Created Arcane-OCR
- Mar 21, 2026Created Arcane-GLM
- Mar 21, 2026Created Arcane-PP
- Mar 25, 2026Created Hailo-Training
- Apr 17, 2026Created Information-Lab
- Apr 24, 2026Most recent push to Information-Lab
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.