01 · Roasts
Hello, World! Simulator
Your 'blazingly fast personal assistant' in `assist` literally contains `println!("Hello, world!")` and nothing else. You wrote more documentation for *a different project* in PROTOTYPE_SUMMARY.md than you wrote actual Rust code.
Repo Graveyard Curator
41 public repos, a 0.24 stale ratio, and two repos (`3d-chess`, `minival`) that are basically just `git init` with a license file. The ideas are clearly flowing — it's the execution that's gasping for air.
127 PRs, 21 Stars
You opened 127 pull requests this year but your entire public portfolio has accumulated 21 stars lifetime. You're doing everyone else's laundry while your own house is on fire.
Quantum Academic Flexer
Two repos (`mos-quantum-learning` + `cs310-final-report`) dedicated to the same quantum learning paper, both with 0–1 stars. Caro et al. has more citations than your implementation has forks.
Heatmap Intermittent Faster
Weeks 1–6, 10–13, 22, 25, 33–35 of your heatmap: all zeros. With 350 commits squeezed into the remaining weeks, your GitHub activity looks less like a career and more like a series of caffeine emergencies.
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% weight80A
- Community10% weight50D
03 · Stats
365-day commit heatmap
89 active days
Language distribution
- Python36%
- C++20%
- Rust13%
- JavaScript7%
- TeX6%
- Typst4%
- Other14%
04 · Numbers
Owned repos
non-fork
38
Commits
last 12 months
350
Followers
52
Joined GitHub
Jul 2022
05 · Top repos
a1exxd0 /
mos-quantum-learning
Rigorous academic implementation of Caro et al.'s quantum learning protocol with comprehensive test suite, CI, and documented architecture; 19MB codebase with 30+ experiments spanning proof completeness, soundness, and robustness.
a1exxd0 /
uow-report-template
Typst-based University of Warwick report template with dual layouts (report & problem-set), theorem environments, and quantum circuit examples. Documented, structured, and actively maintained with 11 commits over 2+ months.
a1exxd0 /
alex-a-prototype
Personal Discord bot assistant with LLM tool-calling, semantic memory (LanceDB), and modular skills (Gmail, Calendar). Typed Python with comprehensive tests and CI, clear architecture, but minimal adoption (1 star, 0 forks, explicitly not for external use).
a1exxd0 /
cs324
CS324 coursework: a JavaScript/Three.js 3D game with two playable levels, state management, collision detection, and interactive objects. Typed structure and meaningful documentation, but no tests or CI.
a1exxd0 /
cs310-final-report
Academic final-year project report on quantum learning protocol implementation using Typst. Well-documented research paper with companion implementation, recent activity (28 of 30 commits), but minimal stars/adoption and limited project scope.
a1exxd0 /
nvim
Personal neovim configuration using LazyVim framework. Basic setup with plugin customization for Rust/C++/Python development, 25KB codebase, minimal documentation beyond brief README.
a1exxd0 /
cs324-report
University of Warwick LaTeX report template with basic structure (chapters/, utils/preamble.tex, Makefile). No stars, fresh repo (9 days old), 5 commits across brief window. Typed language (TeX) with README and .gitignore but no license or CI.
a1exxd0 /
assist
Early-stage Rust personal assistant with placeholder main.rs, no tests, no CI, no documentation, 3 commits in 2 days. GPL-3.0 licensed but lacks meaningful implementation beyond scaffolding.
a1exxd0 /
minival
Empty scaffold created 2026-03-02 with only 2 commits, 15 KB total size, no code files sampled, no README, no tests, no CI. Appears to be a one-shot initialization dump with GPL-3.0 license and .gitignore but no actual implementation.
a1exxd0 /
3d-chess
Empty scaffold with no files, no documentation, and no commits since creation. Repo created and pushed same moment (2026-03-14) with zero recent activity and no source code to evaluate.
06 · Timeline
- Jul 17, 2022Joined GitHub
- Oct 28, 2025Created nvim — my neovim config
- Jan 8, 2026Created cs324 — cs324 computer graphics coursework - arctic research station
- Jan 20, 2026Created mos-quantum-learning — A thorough implementation of Caro et al.'s verified quantum learning protocol. (https://arxiv.org/abs/2306.04843)
- Jan 23, 2026Created uow-report-template — Report template for LaTeX submissions to the University of Warwick, made with Typst.
- Jan 23, 2026Created cs324-report
- Feb 25, 2026Created alex-a-prototype — my personal assistant
- Mar 2, 2026Created minival — cutting down eval time systematically for agentic testing
- Mar 11, 2026Created assist — your blazingly fast personal assistant on linux/windows/macos
- Mar 14, 2026Created 3d-chess — perspective tracking 3d chess game
- Apr 4, 2026Created cs310-final-report — Implementing and evaluation a protocol for classically verifiable quantum learning.
- Apr 13, 2026Most recent push to uow-report-template
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.