01 · Roasts
Hibernation Champion
22 consecutive empty weeks on your heatmap, then frantic burst-mode coding. Your GitHub looks like a bear that wakes up every few months to build a 50k-LOC Rust engine.
2 Followers, Infinity Ambition
You've shipped a Chrome extension with pgTAP database tests, a top-5 Citadel terminal bot, and a globally-ranked trading system — and somehow convinced exactly 2 people to follow you. Marketing budget: $0.
Test Coverage: Selective
zetalog has Playwright E2E running real Chromium 3x + pgTAP DB tests. foodbank-optimisation has... 200 lines in one function and vibes. The variance on this profile is genuinely alarming.
Solo Artist, No Features
soloPct = 100%. Every single commit, every repo, zero collaborators. CitadalTerminal beat Georgia Tech's team. You're out here playing 1v5 and calling it a fair fight.
The Profile README
TahaKhanM repo scored a 20 — the lowest on your entire profile. Your personal brand repo is your weakest shipped artifact. The irony is not lost.
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% weight68C
- Consistency20% weight65C
- Quality20% weight77B
- Depth15% weight70B
- Breadth10% weight65C
- Community10% weight30F
03 · Stats
365-day commit heatmap
57 active days
Language distribution
- Python52%
- Jupyter Notebook28%
- TypeScript14%
- Rust2%
- HTML1%
- JavaScript1%
- Other2%
04 · Numbers
Owned repos
non-fork
11
Commits
last 12 months
978
Followers
2
Joined GitHub
Oct 2020
05 · Top repos
TahaKhanM /
CitadelTerminal
Production-grade competition AI with ~50k LOC: Rust simulator engine reverse-engineered from bytecode + Python search loop. Top-5 placing in 1,000+ competitor tournament, comprehensive documentation, no tests/CI but typed Rust/Python code with architectural clarity.
TahaKhanM /
Prosperity
Prosperity 4 IMC trading competition workspace: 329 MB Rust/Python backtester, options pricing suite, market-making traders, counterparty analysis. Team peaked #32 globally. Well-documented (design.md, ARCHITECTURE.md, STATUS.md) but no tests/CI/license.
TahaKhanM /
zetalog
ZetaLog: a Chrome extension + web leaderboard for Zetamac mental-arithmetic game. Validates game telemetry server-side with rigorous anti-cheat (consistency, physiology, statistical problem-stream checks), uses typed Zod schemas throughout, ships with tests (HAS_TESTS=yes), CI (HAS_CI=yes), and extensive docs (design.m
TahaKhanM /
AI-Agent-Hackathon
Precedent: Enterprise incident-resolution agent with deterministic policy engine, ACL-gated knowledge retrieval, typed tool execution, and hash-chained audit. Measured against real ServiceNow data (94.4% fix-class repeatability). Comprehensive architecture, tests, and CI—but nascent (3 weeks old, 0 GitHub stars).
TahaKhanM /
FoundersHQ
Full-stack fintech MVP with deterministic financial core, evidence-linked insights, and LLM guardrails. Hackathon winner with typed Python backend, CI/tests, but 0 stars and pre-launch product focus.
TahaKhanM /
ascendra
Gamified career-prep platform with multi-judge AI scoring pipeline and Elo ratings. FastAPI backend (Python, no type hints) + Next.js frontend, structured 14 bounded contexts, comprehensive docs, 13 commits over 3 months. Active portfolio project shipped beyond hackathon MVP.
TahaKhanM /
led-panel-pong-emulator
Educational coursework project: C Pong game + faithful browser emulator for 32×32 RGB LED panel protocol. Well-documented architecture with HAL abstraction, Emscripten integration, and debug features; unfinished game logic and no tests/CI limit production readiness.
TahaKhanM /
neural-network-from-scratch
Educational NumPy neural network on MNIST in single Python file with comprehensive mathematical documentation in README, but lacks tests, CI, types, and production structure.
TahaKhanM /
foodbank-optimisation
Academic research implementation using PuLP to solve MILP for nutritionally constrained, cost-minimized food bank parcels. Code is functional but minimally documented, lacks tests/CI, and has shallow version control history (2 of last 30 commits).
TahaKhanM /
microfinance-preprint-repo
Academic research preprint repository with pandas/scikit-learn analysis of microfinance impact on female employment in India. Python scripts directly transcribed from notebooks; no tests, CI, or typed code. Structured for reproducibility but thin on engineering practices.
TahaKhanM /
TahaKhanM
GitHub profile README showcasing CS student's technical interests and skills. Purely informational portfolio page with no code, no examples, no structure—a one-off placeholder repo.
06 · Timeline
- Oct 31, 2020Joined GitHub
- Sep 8, 2025Created neural-network-from-scratch — NumPy neural network from scratch with SGD/backpropagation on MNIST.
- Feb 6, 2026Created microfinance-preprint-repo — pandas/scikit-learn analysis of microfinance, MFI operations, and women's employment outcomes.
- Feb 6, 2026Created foodbank-optimisation — Linear-programming optimisation model for nutritionally constrained, lower-cost food bank parcels.
- Feb 6, 2026Created led-panel-pong-emulator — C Pong game and browser emulator for a 32x32 RGB LED panel protocol.
- Feb 21, 2026Created TahaKhanM — GitHub profile README for Taha Khan's technical portfolio.
- Feb 28, 2026Created FoundersHQ — Full-stack fintech platform for startup invoices, runway forecasting, search, REST APIs, and OpenAPI docs.
- Mar 28, 2026Created ascendra — Gamified, competitive career-prep & candidate-benchmarking platform — pathway arenas, calibrated multi-judge AI scoring, Elo ratings, skill tree, and replays. FastAPI + Next.js.
- Apr 12, 2026Created Prosperity — IMC Prosperity trading workspace: options pricing, market making, statistical arbitrage, Rust/Python backtesting.
- Apr 23, 2026Created CitadelTerminal — Citadel Terminal AI challenge workspace with Python strategies, replay analysis, and Rust simulation tooling.
- Jul 2, 2026Created AI-Agent-Hackathon
- Aug 9, 2026Created zetalog
- Aug 17, 2026Most recent push to zetalog
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.