01 · Roasts
Speed-run Coder
verseLock: entire Kotlin Android app committed in 19 minutes. RobStatTM-py: born and completed in 11 minutes. multilingual-docs: cradle-to-grave in 44 minutes. Are you coding or speed-running?
CI? Never Heard of Her
Zero CI pipelines across all 10 analyzed repos. You've written 40+ tests in RobStatTM-py but apparently trust vibes over automation to run them.
The Zero-Star Constellation
28 public repos. 0 total stars. 0 total forks. You're shipping into a black hole — the universe hasn't noticed yet.
HTML/Jupyter Dichotomy
Your language breakdown is 49% HTML and 48% Jupyter Notebook. Your actual code — TypeScript, Kotlin, Python — rounds to 1%. The byte counter is telling on you.
GSoC Grind Machine
multilingual-docs, ecotourism-tests, foa-pipeline — at least three repos are explicitly GSoC applications. Portfolio-building is a skill; shipping one for real would be a flex.
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% weight55D
- Quality20% weight57D
- Depth15% weight50D
- Breadth10% weight55D
- Community10% weight55D
03 · Stats
365-day commit heatmap
64 active days
Language distribution
- HTML49%
- Jupyter Notebook48%
- Python1%
- Kotlin0%
- TypeScript0%
- JavaScript0%
- Other2%
04 · Numbers
Owned repos
non-fork
13
Commits
last 12 months
99
Followers
14
Joined GitHub
Apr 2020
05 · Top repos
adit-0132 /
llm-context-router
Browser extension for exporting LLM conversations between ChatGPT and Claude. Well-scoped, functional implementation with typed manifest, structured file layout, and documented .llmchat format. Shipped project with meaningful use case but limited external adoption (0 stars, single author).
adit-0132 /
tds-virtual-rag
Educational RAG system for IIT Madras TDS course with multimodal CLIP embeddings, ChromaDB vector search, web scraping, and dual LLM fallback. Typed Python with structured architecture, README, and working code—but experimental stage with zero adoption signals and minimal commit activity.
adit-0132 /
ecotourism-tests
A Shiny app + utility functions for exploring Australian wildlife sightings, built for GSoC 2026 selection. Well-designed R project with typed code, structured layout, clear docs, and interactive visualizations, but no tests or CI, and limited adoption signals.
adit-0132 /
RobStatTM-py
Early-stage Python wrapper for R's RobStatTM robust statistics package. Minimal adoption (0 stars), but functional typed code with comprehensive docstrings and test suite. Just 5 commits in under 30 minutes; experimental rather than sustained.
adit-0132 /
verseLock
Kotlin Android app with polished Jetpack Compose UI, Room database caching, and Ktor networking. Well-documented architecture and privacy-first design, but extremely fresh (3 hours old, 3 commits), no tests, no CI, and zero adoption signals yet.
adit-0132 /
foa-pipeline
A Python web-scraping pipeline for federal funding opportunities with rule-based semantic tagging. Typed-language, multi-file structure, meaningful README, but experimental single-creator effort with 44 KB codebase, 8 commits in 6 hours, no tests/CI/license.
adit-0132 /
copilot-router
Early-stage VS Code extension (v0.1.0, 1 commit in 4 minutes) with TypeScript + tests + structured architecture that intelligently routes Copilot queries to cheaper models via hybrid heuristic/LLM classifier to save API budget.
adit-0132 /
multilingual-docs
GSoC demonstration R package with dynamic Rd documentation via Sexpr macros. Single exported function, 8 tests, MIT license, created and pushed within same day (2026-03-29).
adit-0132 /
machine-learning-practice
Minimal one-day Jupyter Notebook practice repo with empty README, no tests/CI, and sparse commit activity (7 of 30 days). Appears to be a personal learning dump with no discernible project scope or documentation.
adit-0132 /
my-pages
Empty scaffold repo with one-shot commit, no docs, no structure. Personal experiment with no adoption or documented purpose.
06 · Timeline
- Apr 21, 2020Joined GitHub
- Jun 16, 2025Created tds-virtual-rag
- Jan 23, 2026Created llm-context-router
- Jan 31, 2026Created machine-learning-practice
- Feb 26, 2026Created my-pages
- Mar 1, 2026Created foa-pipeline — FOA Ingestion + Semantic Tagging Pipeline — fetches Grants.gov and NSF funding opportunities via API and HTML scraping, tags them against a controlled keyword ontology, and exports
- Mar 4, 2026Created ecotourism-tests
- Mar 16, 2026Created verseLock
- Mar 24, 2026Created RobStatTM-py — robStatTM Robust Statistics made available via RPy2 wrappers to python
- Mar 29, 2026Created multilingual-docs — An R package demonstrating dynamic Rd documentation via \Sexpr macros, for the Multilingual Documentation of R Packages project.
- Mar 31, 2026Created copilot-router — a VS Code Extension that routes AI agent queries based on complexity to cheapest sufficient model available, and can save tonne of AI creds.
- Mar 31, 2026Most recent push to copilot-router
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.