01 · Roasts
The Graveyard Architect
You named a repo 'medcify-plus' and then never committed a single file. It's not a project — it's a GitHub-shaped empty box. Even your ambitions are stubs.
Sprint King, Maintenance Peasant
civic-os: 30 commits in 18 days then silence. arenaflow: 3 days old. medcify-y: still going. You build like a hackathon is always ending tomorrow — shipping streaks great, but none of these has seen a second month.
65% Notebook, 0% Tests
Your repo bytes are 65% Jupyter Notebook and you have exactly one repo with tests (vitest in arenaflow, written on day 1 and not touched since). A Jupyter cell is not a unit test, Bishal.
The AI Noun Collector
Gemini AI, Sarvam AI, LLM OCR, Zebra AI, intervue-ai — you've successfully named every OpenAI competitor in your READMEs. Zero CI pipelines across 9 repos. The LLMs are load-bearing; the engineering scaffolding is not.
3 PRs, 0 Issues, 96% Solo
totalPRsYear = 3, totalIssuesYear = 0, soloPct = 96%. You are building in a vacuum so complete that even your own repos haven't filed issues against each other.
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% weight56D
- Consistency20% weight55D
- Quality20% weight57D
- Depth15% weight50D
- Breadth10% weight65C
- Community10% weight25F
03 · Stats
365-day commit heatmap
46 active days
Language distribution
- Jupyter Notebook65%
- TypeScript31%
- HTML2%
- Python1%
- JavaScript1%
- CSS0%
04 · Numbers
Owned repos
non-fork
18
Commits
last 12 months
124
Followers
9
Joined GitHub
Aug 2024
05 · Top repos
Bishu-21 /
civic-os
TypeScript civic complaint platform with Next.js 16, Appwrite backend, Gemini AI analysis, voice interface via Sarvam, Leaflet mapping, and server-side security (sanitization, rate limiting, RLS). Active 18-day sprint with 30 commits, comprehensive architecture docs, but no tests/CI and missing production hardening.
Bishu-21 /
medcify-y
Ambitious AI-driven pharmacy/healthcare platform combining OCR, predictive analytics, and multimodal LLMs. Typed (TS), documented README, structured src/ layout with server actions and lib services. No tests, CI, or license. ~2 months active, 14 of last 30 commits, ~1300 KB indicates real scope. Personal portfolio proj
Bishu-21 /
zebra-ai
Zebra AI is a TypeScript-based resume optimization SaaS platform featuring AI-powered analysis, job tailoring, and PDF export. Typed, documented, and architecturally sound, but nascent: 1 star, 8 recent commits, and 2-week lifespan with no external adoption signals.
Bishu-21 /
my-openenv
Educational OpenEnv support-triage environment with typed models, deterministic graders, and FastAPI backend. Personal project in very early stage (3 days old, 14 commits), narrow scope focused on single domain.
Bishu-21 /
arenaflow
Early-stage Next.js stadium crowd-management platform with Gemini AI integration, Neon Postgres, and Razorpay billing. TypeScript + tests present; minimal stars/adoption but demonstrates structured, typed architecture.
Bishu-21 /
zebra
Early-stage resume AI SaaS monorepo (Zebra) with TypeScript frontend, FastAPI worker, and Drizzle ORM schema. Typed, documented with README and turbo.json setup, but just 3 recent commits across <1 day; no tests, CI, or license; minimal architectural depth for claimed "production-ready" claim.
Bishu-21 /
intervue-ai
Minimal Next.js scaffold for an AI interview platform; created and last pushed same day with 4 commits; no tests, CI, or meaningful project documentation beyond boilerplate Next.js README.
Bishu-21 /
Bishu-21
Profile config repository with README only. Zero stars, no code artifacts sampled, minimal commits (3 of last 30), no tests/CI/license. Purely personal biographical content.
Bishu-21 /
medcify-plus
Empty scaffold created 2026-02-20 with zero commits, zero files, no README, tests, CI, or documentation. Appears to be an unpushed repository placeholder.
06 · Timeline
- Aug 8, 2024Joined GitHub
- Sep 29, 2024Created Bishu-21 — Config files for my GitHub profile.
- Feb 20, 2026Created medcify-plus
- Feb 20, 2026Created medcify-y
- Feb 28, 2026Created zebra
- Mar 10, 2026Created civic-os — An AI-powered civic complaint management system for Delhi citizens and municipal officers.
- Mar 11, 2026Created intervue-ai — A real-time AI Avatar Interview Platform where a candidate gives one intelligent interview conducted by an AI avatar.
- Apr 4, 2026Created zebra-ai
- Apr 8, 2026Created my-openenv
- Apr 19, 2026Created arenaflow
- Apr 22, 2026Most recent push to medcify-y
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.