01 · Roasts
The 48-Hour Graveyard
RAG-AI-assistant and Streaming-Video-Platform were born and abandoned within a single day. Two repos, two commits each, both scored under 30. At this rate your GitHub is less a portfolio and more a startup cemetery.
Test? Never Heard of Her
Out of 10 repos, exactly one has real tests — and even those are placeholder comments saying 'Will be fully implemented in Phase 3.' Phase 3 has never arrived in recorded history.
95% Python, 0% pytest
Python is 95% of your codebase by bytes. You clearly love the language. You just refuse to write a single passing test in it. That's a special kind of commitment to living dangerously.
Hackathon Hero, CI Zero
WardWatch won a hackathon with Elasticsearch + AWS Bedrock built in 4 hours. Genuinely impressive. It also has no CI, no tests, and no license. Civic tech deserves better than cowboy deployments.
Burst Mode Developer
Your heatmap is a horror movie: 30 consecutive weeks of near-silence, then a frantic daily burst in the last 5 weeks. The year doesn't have a commit history — it has a plot twist.
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% weight65C
- Depth15% weight58D
- Breadth10% weight55D
- Community10% weight55D
03 · Stats
365-day commit heatmap
75 active days
Language distribution
- Python95%
- C2%
- Cython1%
- TypeScript1%
- JavaScript0%
- C++0%
- Other1%
04 · Numbers
Owned repos
non-fork
57
Commits
last 12 months
368
Followers
39
Joined GitHub
Jan 2022
05 · Top repos
chemicoholic21 /
Git-Pull-Talent
Functional Next.js tool for GitHub contributor impact scoring with typed backend, polished UI, multi-tier caching (Redis + Postgres), and public deployment. 20 stars, active development, no tests/CI, complete feature set with leaderboards and user profiles.
chemicoholic21 /
github-data-pipeline
A TypeScript headless pipeline for discovering GitHub developers, scoring contributions (stars × userPRs/totalPRs), and storing ranked profiles in Postgres. Typed, documented, structured with scoring logic and multi-stage architecture, but experimental scope (0 stars) and early-stage development limit impact.
chemicoholic21 /
Portfolio-website
Personal portfolio site built with Next.js + TypeScript, featuring GSAP animations and custom React components. Typed language, structured layout, README present; lacks tests/CI/license but demonstrates solid foundational craftsmanship.
chemicoholic21 /
WardWatch
Hackathon-winning civic observability platform for Bengaluru using Elasticsearch, AWS Bedrock, and TypeScript. Detects unresponsive gov offices, analyzes scam messages, traces complaint escalations. Built solo in ~4 hours with structured backend APIs but thin frontend presence and no tests/CI.
chemicoholic21 /
PRISM
Early-stage TypeScript/Python full-stack app (17 MB) for detecting PR intent drift using local Gemma inference. Typed, documented, and architecturally structured, but pre-launch with minimal adoption signals.
chemicoholic21 /
T-Rex
Console-based Chrome T-Rex game clone in C++ with Python variant. Minimal scope (48 KB) with jumping, scoring, and rendering. README present with build/play instructions. No tests, no license, no gitignore despite CI setup and 3-year development window (2023–2026).
chemicoholic21 /
Travel-Planner
Personal Flask travel planner with AI itinerary generation and weather integration. Typed language missing, no tests/CI, but README present, working implementation with user auth and basic structure.
chemicoholic21 /
RAG-AI-assisstant
Early-stage RAG assistant with well-documented architecture and thoughtful security design, but unproven and experimental. Fresh repo (48 hours old, 2 commits) with incomplete test suite and no CI pipeline. Promising craft in ingestion/retrieval logic, but thin execution overall.
chemicoholic21 /
Streaming-Video-Platform
Very early-stage video streaming platform (1 day old, 2 of 30 commits fetched) with minimal documentation, no tests visible in samples, no CI/license, CSS primary language, and zero community adoption.
chemicoholic21 /
chemicoholic21
Personal portfolio README with no actual code repository—just a CV/resume listing tech skills and hackathon wins. No source files, no project artifacts, no meaningful codebase.
06 · Timeline
- Jan 6, 2022Joined GitHub
- Feb 19, 2023Created T-Rex — Chrome t-rex game in c++
- Sep 26, 2023Created chemicoholic21
- Jul 14, 2024Created Travel-Planner
- Jul 17, 2025Created Portfolio-website
- Mar 6, 2026Created Git-Pull-Talent — A public web tool that analyzes a GitHub user’s open-source contributions and computes meaningful impact scores based on repository popularity and pull-request activity.
- Mar 31, 2026Created github-data-pipeline — A headless data pipeline that ingests, caches, and analyzes GitHub data, transforming it into structured insights stored in Postgres and Redis
- Apr 14, 2026Created Streaming-Video-Platform — Full-stack video upload, processing, sensitivity analysis, and streaming platform with real-time progress tracking.
- Apr 28, 2026Created RAG-AI-assisstant — Domain-specific AI assistant built using a custom RAG pipeline that delivers accurate, context-grounded responses while enforcing strict domain boundaries and defending against pro
- May 9, 2026Created WardWatch — Civic Navigator (featured OpenClaw) - OpenClaw, Bedrock AgentCore EC2, Elasticsearch, Open Crawler, Agent Builder, MCP
- May 16, 2026Created PRISM — Pull Request Intelligence & Security Machine
- May 29, 2026Most recent push to WardWatch
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.