01 · Roasts
The 3-Day Founder
ProdReady has full CI, 16 rule modules, Prisma migrations, and a vitest suite — built in 72 hours. Either you have a time machine or you're optimizing for GitHub activity graphs, not users. It has 2 stars; one of them is probably you.
Test Allergy
10 out of 12 repos have HAS_TESTS=no. You built a CNN brain tumor detector (NeuroScan-AI), a 7-algorithm visualizer (Algoviz), and a 3D VR game (UpsideDown) — all untested. That's not confidence, that's vibes-driven deployment.
Solo Artist, No Bandmates
soloPct=99% across 32 repos. 347 commits this year and not a single meaningful collaboration. Your code has never had to survive contact with another developer's opinion.
Heatmap Hibernator
Your contribution heatmap looks like a bear woke up in March — 25 weeks of near-silence followed by a frantic burst of 4s and 3s. That's not consistency, that's deadline panic dressed up as productivity.
Star Collector (Struggling Edition)
32 public repos. 12 total stars. That's 0.375 stars per repo. NeuroScan-AI has 0 stars despite a 52 MB codebase and Groq chatbot integration. The market has spoken very quietly.
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% weight72B
- Depth15% weight58D
- Breadth10% weight65C
- Community10% weight55D
03 · Stats
365-day commit heatmap
120 active days
Language distribution
- TypeScript43%
- Jupyter Notebook32%
- HTML14%
- JavaScript6%
- Python3%
- CSS1%
- Other1%
04 · Numbers
Owned repos
non-fork
28
Commits
last 12 months
347
Followers
11
Joined GitHub
Nov 2022
05 · Top repos
nsk6704 /
ProdReady
Production-readiness SaaS tool with Next.js stack, comprehensive scanner engine for 16 OSS rules, full CI/CD, typed codebase, and API endpoints. Early-stage but well-structured indie product.
nsk6704 /
myPortfolio
A personal portfolio built with Next.js and TypeScript, featuring GitHub API integration, custom components, and animations. Well-structured and typed, but thin public impact (0 stars) as a personal project with no external adoption.
nsk6704 /
PlacementCompanion
TypeScript coursework project for mental health/placement tracking with JWT auth, FastAPI backend, Next.js frontend, research-grounded insights engine, and 182 KB codebase built in ~3 months with 23 recent commits.
nsk6704 /
Algoviz
Interactive web-based algorithm visualizer built with Next.js and TypeScript. Covers 7 ML/AI algorithms (K-means, KNN, A*, Minimax, etc.) with interactive canvas-based visualizations, responsive UI, but lacks tests, CI, and comprehensive documentation.
nsk6704 /
Telemetron
Lightweight Go telemetry service for multi-agent AI systems; well-structured with mock repositories, unit tests, and comprehensive docs; early-stage personal project with no production users yet.
nsk6704 /
NeuroScan-AI
Personal full-stack medical imaging project with CNN brain tumor detection, FastAPI backend, vanilla JS frontend, and Groq chatbot integration. Typed Python not evident; no tests or CI; well-documented with clear deployment guide; 52849 KB codebase built across 4+ months suggests substantial scope.
nsk6704 /
InterviewAssistant
Personal TypeScript AI interview coaching app with FastAPI backend, voice I/O via Groq APIs, and polished React UI. Typed, documented, structured codebase lacking tests and CI but demonstrates real product work.
nsk6704 /
UpsideDown
A JavaScript-based VR-ready 3D game built with Three.js featuring a jungle and crystal cave world with interactive elements. 4MB codebase, 15 commits over ~3.5 months, typed language unavailable, no tests or CI, but documented with README and customization guide.
nsk6704 /
UpWell
UpWell is a React Native wellness/finance tracker with TypeScript, SQLite persistence, multi-category tracking, and gamification (XP/levels). ~4 months of active development with 30 commits, comprehensive app architecture across 5 screens, but no tests, CI, or external adoption yet.
nsk6704 /
AnomalyDetect
Personal anomaly detection framework with 5 model architectures (LSTM, CNN, GRU, Transformer, Dense), parallel training engine, and FastAPI+React UI. Untyped Python, no tests, no CI, lacks architectural documentation.
nsk6704 /
upwell-website
Landing page for UpWell Android app; single-star personal project with 1.1 MB of HTML/CSS, 6 commits over 4 months. README documents features and contact; lacks tests, CI, license, and code samples indicate static HTML structure.
nsk6704 /
nsk6704
A portfolio README showcasing the owner's full-stack work (InterviewAssistant, UpWell) with links to external projects, but the repo itself contains no source code—it's a profile landing page with 37 KB total size and no implementation.
06 · Timeline
- Nov 9, 2022Joined GitHub
- Dec 4, 2023Created nsk6704
- Mar 18, 2025Created Algoviz — AlgoViz is an interactive web-based platform designed to visualize algorithms in a clear, step-by-step manner.
- Nov 24, 2025Created InterviewAssistant — An AI-powered interview practice application designed to simulate realistic technical interviews with a focus on conversational quality, adaptability, and professional feedback.
- Dec 21, 2025Created AnomalyDetect — AnomalyDetect is an anomaly detection framework that supports training multiple deep learning models such as LSTM, CNN, Transformer, and GRU in parallel.
- Jan 3, 2026Created UpsideDown — An immersive, multi-level VR-ready treasure hunt game built with Three.js. Explore a lush jungle, discover secret cave entrances, and descend into the glowing Crystal Core.
- Jan 4, 2026Created NeuroScan-AI — NeuroScan AI is a full-stack platform that uses a CNN to detect and classify brain tumors from MRI scans along with a context-sensitive AI Chatbot.
- Jan 15, 2026Created UpWell — Upwell is a cross-platform mobile application designed to help you build lasting healthy routines and financial discipline.
- Jan 15, 2026Created upwell-website — Landing Page for my mobile app UpWell
- Jan 24, 2026Created PlacementCompanion — This is a repository created for the Data and Storytelling Course EL
- Jan 30, 2026Created myPortfolio — For my personal potfolio
- Feb 6, 2026Created Telemetron — Telemetron is a lightweight telemetry service that provides a unified, machine-readable snapshot of system state for multi-agent AI systems.
- May 15, 2026Created ProdReady
- May 17, 2026Most recent push to ProdReady
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.