01 · Roasts
Portfolio, meet users
Ten repositories and zero stars: the products are shipping faster than the audience is arriving.
Test suite has a favorite
gradesense brings 161 tests and jobagent is well covered; several other repos still ship with the placeholder test script.
CI is a rare Pokémon
AI_Code_Review_Tool has CI, while the rest of this ten-repo portfolio mostly relies on vibes and local runs.
Architecture maxxing
Search-TypeAhead-System assembles Kafka, PostgreSQL, and six Redis nodes in one commit—distributed systems any% speedrun.
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% weight65C
- Depth15% weight55D
- Breadth10% weight55D
- Community10% weight25F
03 · Stats
365-day commit heatmap
28 active days
Language distribution
- TypeScript89%
- JavaScript3%
- Java3%
- Shell1%
- Swift1%
- CSS1%
- Other2%
04 · Numbers
Owned repos
non-fork
10
Commits
last 12 months
82
Followers
1
Joined GitHub
Jul 2026
05 · Top repos
31UTKARSH07-ALPHA /
jobagent
A documented, typed TypeScript job-application pipeline with SQLite state, ATS/Gmail ingestion, LLM scoring, guarded outreach, and substantial tests; it has no stars or external-adoption evidence yet.
31UTKARSH07-ALPHA /
gradesense
GradeSense is a documented, typed TypeScript grading pipeline with evidence verification, OCR-aware abstention, editable PDF annotations, SQLite history, and a substantial offline test suite; adoption is not yet demonstrated.
31UTKARSH07-ALPHA /
AI_Code_Review_Tool
Typed TypeScript Express tool with CLI, REST API, Prisma/PostgreSQL, Redis caching, Docker Compose, and GitHub webhook handling; documented and CI-built, but unadopted and lacking tests.
31UTKARSH07-ALPHA /
Notebook-RAG
A documented, typed Next.js RAG application with PDF/TXT ingestion, Qdrant hybrid retrieval, reranking, grounded generation, and an evaluation dashboard, but it is a zero-star one-day project without tests, CI, or license.
31UTKARSH07-ALPHA /
Learn_DevOps
A well-documented seven-part DevOps coursework repository with reproducible shell labs, Dockerfiles, networking demonstrations, and captured outputs, but no visible adoption, tests, CI, license, or typed implementation.
31UTKARSH07-ALPHA /
Distributed-Cache-System
A documented, typed Java cache demo with LRU, TTL, metrics, events, and consistent hashing, but no tests, CI, license, adoption signals, or evidence of sustained repository evolution.
31UTKARSH07-ALPHA /
Search-TypeAhead-System
A documented, typed TypeScript typeahead system with Kafka batching, PostgreSQL indexing, Redis Cluster serving, consistent hashing, ranking strategies, and a React search UI, but currently a zero-star one-commit project without tests or CI.
31UTKARSH07-ALPHA /
customer-support-agent
A small TypeScript customer-support-agent scaffold with a health endpoint and partially implemented RAG ingestion/embedding modules, but no documented product surface, tests, CI, or demonstrated adoption.
31UTKARSH07-ALPHA /
website-automation-agent
A documented JavaScript Playwright/LLM browser-agent prototype with a 15-iteration loop, modular tools, screenshots, and logging, but no tests, CI, license, or demonstrated adoption.
31UTKARSH07-ALPHA /
31UTKARSH07-ALPHA
A newly created profile README repository with no sampled source files, tests, CI, license, or typed implementation; it presents personal technology interests but no demonstrated software artifact.
06 · Timeline
- Jul 8, 2026Joined GitHub
- Jul 8, 2026Created Notebook-RAG
- Jul 8, 2026Created Search-TypeAhead-System
- Jul 8, 2026Created Distributed-Cache-System
- Jul 8, 2026Created website-automation-agent
- Jul 11, 2026Created AI_Code_Review_Tool
- Aug 6, 2026Created customer-support-agent
- Aug 9, 2026Created jobagent
- Aug 31, 2026Created 31UTKARSH07-ALPHA
- Sep 1, 2026Created gradesense — Rubric-based grading and annotation tool: verified evidence, honest abstention on unreadable diagrams, editable annotations, annotated PDF export.
- Sep 3, 2026Created Learn_DevOps — DevOps class assignments - Utkarsh Pathak (24bcs10309). All 7 assignments with real captured command output, screenshots and explanations.
- Sep 3, 2026Most recent push to jobagent
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.