01 · Roasts
The 30-Minute SaaS Founder
Malik-SkillSync went from zero to 'live SaaS product with Railway endpoints and a .lk domain' in exactly 30 minutes. That's not shipping — that's a git push cosplaying as a startup launch.
The Burst Artisan
7 of your 10 repos were created and last pushed on the same day. NeuralEye lived for 4 minutes. SixtySevenDetector lasted 1 minute. HandCanvas squeezed 30 commits into 3 days. You don't build software — you speed-run it.
The 0-Test Absolutist
10 repos. 0 tests. 0 CI pipelines. Not a single pytest, Jest, or GitHub Action in sight. Somewhere a QA engineer just felt a disturbance in the Force.
Heatmap Ghost
Your heatmap is basically 40 weeks of tumbleweeds followed by a panicked sprint in the last 2 months. Either you discovered GitHub in March 2026 or your professor set a deadline.
Portfolio Inflation Detected
Two repos called SkillSync (SkillSync + Malik-SkillSync), same stack, same concept, zero forks between them. Duplicating your own projects doesn't double your impact — it halves your credibility.
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% weight59D
- Depth15% weight58D
- Breadth10% weight65C
- Community10% weight55D
03 · Stats
365-day commit heatmap
43 active days
Language distribution
- TypeScript73%
- Python24%
- Java1%
- CSS1%
- HTML0%
- JavaScript0%
- Other1%
04 · Numbers
Owned repos
non-fork
18
Commits
last 12 months
391
Followers
14
Joined GitHub
Jul 2024
05 · Top repos
malik-builds /
SkillSync
SkillSync is a full-stack TypeScript/Python talent alignment platform with AI-powered CV analysis, GitHub auditing, and job matching. Ships with 30 commits over ~7 weeks, structured backend (FastAPI + MongoDB + LangGraph) and frontend (Next.js), but lacks tests, CI, and license.
malik-builds /
HealthLens
Full-stack TypeScript diabetes risk assessment app: Next.js 15 frontend with Three.js helix, FastAPI backend with RandomForest model on Pima dataset. Typed, documented, structured, no CI/tests; young repo (18 days) with 11 of last 30 commits.
malik-builds /
HandCanvas
TypeScript/React gesture-drawing app using MediaPipe Hands via CDN. Fully functional with TypeScript, structured components, and comprehensive README. No tests, CI, or license; minimal commits (30) in 3-day window suggests burst work rather than sustained development.
malik-builds /
ApexPredict
Full-stack F1 race predictor (RandomForest+XGBoost ensemble + FastAPI + Next.js) with educational walk-through intent. Untyped Python, clean API, typed frontend, but minimal adoption and very recent.
malik-builds /
5COSC022C.2-CSA-CW-UoW
University coursework: a 3-layer Java JAX-RS REST API for campus sensor management with proper service/DAO structure, synchronization, exception handling, and documented API design—but limited scope (academic submission, no external adoption)
malik-builds /
better-llm-council
Claude skill implementing Karpathy's LLM Council via sub-agents for structured decision-making. Well-documented concept with clear examples, but extremely early-stage: created Apr 27 2026, minimal commits (3 of 30), no tests/CI, untyped language, 13kb total size.
malik-builds /
Malik-SkillSync
TypeScript SaaS platform for AI-powered career matching in Sri Lanka. Has README with live demo URLs and API docs, structured full-stack architecture (Next.js + FastAPI), but created hours ago (2026-03-28), 30 commits in single burst, no tests/CI/license, unverifiable production claims.
malik-builds /
NeuralEye
Young experimental dual-mode CV app (pose + animal detection via YOLO26) with Flask WebSocket streaming; untyped Python, no tests/CI/license, minimal commit history (10 of 30), created April 2026.
malik-builds /
Pose-Analytics-Suite
Untyped Python pose estimation tool with modular structure, Flask web UI, and Docker support. Early-stage project (7 commits over 6 weeks, 8 KB repo) with README but no tests, CI, or actual source files sampled.
malik-builds /
SixtySevenDetector
One-day real-time gesture detection prototype using YOLO-Pose. Minimal commit history (1 of 30), no tests/CI, untyped Python, but functional implementation with README and structured module layout.
06 · Timeline
- Jul 4, 2024Joined GitHub
- Feb 15, 2026Created Pose-Analytics-Suite — A modular Python suite for real-time human pose estimation and automated gesture alerts using YOLOv26 and OpenCV.
- Feb 15, 2026Created SixtySevenDetector — Real-time gesture detection system using YOLO-Pose to trigger multimedia responses and track performance analytics.
- Mar 9, 2026Created SkillSync
- Mar 28, 2026Created Malik-SkillSync — My Version of SkillSync
- Apr 6, 2026Created ApexPredict — F1 Race Outcome Predictor
- Apr 7, 2026Created NeuralEye — Real-time dual-mode computer vision, human pose estimation & animal detection powered by YOLO26, streamed live.
- Apr 9, 2026Created HealthLens — AI-powered diabetes risk assessment using clinical biomarkersclear, calm, and built for understanding (not replacing) your care team.
- Apr 22, 2026Created 5COSC022C.2-CSA-CW-UoW — CSA CW
- Apr 23, 2026Created HandCanvas — Browser-based drawing app where the user draws on a full-screen HTML5 canvas using webcam hand gestures (MediaPipe Hands via CDN).
- Apr 27, 2026Created better-llm-council — A Claude skill that pressure-tests decisions through independent AI advisors with peer review. Based on Karpathy's LLM Council.
- Apr 27, 2026Most recent push to better-llm-council
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.