01 · Roasts
Docker, then dice roll
darkweb-attribution provisions PostgreSQL, Neo4j, seed services, FastAPI, ML, and a graph UI—then skips tests and CI.
Notebook nation
71% of the language mix is Jupyter Notebook: the models get experiments, but the repos rarely get production guardrails.
Deployment without a crowd
self-healing-scraper has a Render URL and a real pytest suite, yet it still has 0 stars and 0 forks.
Portfolio has receipts
There are 11 public repos and 111 recent sampled commits, but only 3 total stars: shipping is ahead of discoverability.
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% weight63C
- Consistency20% weight60C
- Quality20% weight52D
- Depth15% weight55D
- Breadth10% weight55D
- Community10% weight25F
03 · Stats
365-day commit heatmap
45 active days
Language distribution
- Jupyter Notebook71%
- Python15%
- JavaScript6%
- CSS6%
- HTML2%
- Dockerfile0%
04 · Numbers
Owned repos
non-fork
11
Commits
last 12 months
141
Followers
1
Joined GitHub
Nov 2025
05 · Top repos
KhushiBidhuri22 /
self-healing-scraper
A named, deployed FastAPI self-healing scraper with selector repair, replay validation, health reporting, chaos fixtures, and a meaningful pytest suite, but no visible adoption, CI, license, or static typing.
KhushiBidhuri22 /
darkweb-attribution
A substantial but early-stage dark-web attribution MVP combining FastAPI, PostgreSQL, Neo4j, ML fallbacks, CSV loaders, Docker services, and an interactive frontend; adoption is currently unshown and integration correctness remains uneven.
KhushiBidhuri22 /
phishing-email-detector
A documented phishing-email ML demo combining TF-IDF and 13 metadata features with a Flask UI, but it has only 1 star, no tests or CI, and limited evidence of production adoption.
KhushiBidhuri22 /
fake-news-detector
A documented Jupyter Notebook project that trains and compares four NLP classifiers on 44,898 articles, but has minimal adoption and no tests, CI, license, or deployable application surface.
KhushiBidhuri22 /
KhushiBidhuri22
A polished GitHub profile README presenting three named ML projects and a Python/JavaScript/C++ learning focus, but this repository contains no implementation, tests, CI, license, or adoption signals.
KhushiBidhuri22 /
Portfolio
A small, polished static portfolio site with responsive-style sections, animated presentation, project links, and a contact form, but no documented adoption, tests, CI, or deployment evidence in the repository metadata.
KhushiBidhuri22 /
student-performance-predictor
A documented, one-notebook student-grade regression project comparing scikit-learn models on 395 records and 32 features, but with minimal repository engineering and no demonstrated adoption.
KhushiBidhuri22 /
house-price-predictor
A documented Jupyter notebook demonstrates California housing prediction through scaling, feature engineering, linear/polynomial/Ridge regression, and a TensorFlow neural network, but remains a small educational project without adoption or engineering scaffolding.
KhushiBidhuri22 /
javascript-tic-tac-toe
A small browser Tic-Tac-Toe implementation with a clear HTML/CSS/JavaScript structure, but limited adoption, no automated validation, and several incomplete or fragile game-state behaviors.
KhushiBidhuri22 /
spiderdev-web-project2026
A small student web-project collection with a portfolio page, billing calculator, result calculator, and Netflix-style CSS button, but no documentation, tests, CI, license, or evidence of adoption.
06 · Timeline
- Nov 8, 2025Joined GitHub
- Jun 17, 2026Created spiderdev-web-project2026
- Jun 21, 2026Created house-price-predictor
- Jun 23, 2026Created javascript-tic-tac-toe
- Jun 30, 2026Created student-performance-predictor
- Jul 1, 2026Created Portfolio
- Jul 1, 2026Created KhushiBidhuri22
- Jul 4, 2026Created fake-news-detector
- Jul 22, 2026Created phishing-email-detector
- Aug 23, 2026Created self-healing-scraper
- Sep 7, 2026Created darkweb-attribution
- Sep 14, 2026Most recent push to darkweb-attribution
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.