01 · Roasts
CI witness protection
All 12 scored repositories report no CI. The build pipeline is apparently an oral tradition.
Test suite: mythical
Voxara, Elowen, Dora, RSSFeed, and Nitro have serious architecture, yet every scored repo reports zero tests.
Shipyard, not storefront
You have 100 public repos and 990 yearly commits, but only 21 total stars: prolific output has not converted into visible adoption.
README roulette
RSSFeed and Nitro contain substantial multi-layer systems, then make users infer the product because neither has a README.
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% weight57D
- Depth15% weight58D
- Breadth10% weight80A
- Community10% weight55D
03 · Stats
365-day commit heatmap
293 active days
Language distribution
- TypeScript92%
- Go2%
- CSS2%
- Jupyter Notebook1%
- Dart1%
- Rust1%
- Other1%
04 · Numbers
Owned repos
non-fork
99
Commits
last 12 months
990
Followers
14
Joined GitHub
Aug 2024
05 · Top repos
MehulxBuilds /
Elowen
A documented, strictly typed TurboRepo AI Telegram application with web, worker, bot, Kafka, Redis, Prisma/Postgres, billing, and model integrations, but only 2 stars and no tests, CI, or license.
MehulxBuilds /
voxara
Voxara is a typed Next.js voice-generation application with Clerk organizations, Prisma/Postgres, UploadThing audio storage, Chatterbox API integration, custom voice recording, and waveform playback, but has only 1 star and lacks tests, CI, and a license.
MehulxBuilds /
dora
Dora is a documented, typed Go embedded key-value database with JSON persistence, WAL recovery, transactions, immutable JSON-like values, and concurrency protection, but it has 1 star and no tests or CI.
MehulxBuilds /
RSSFeed
A structured Go RSS API with chi routing, PostgreSQL migrations, API-key auth, feed following, and a concurrent RSS scraper; it is a substantial implementation but has no documented usage, tests, CI, license, or visible adoption.
MehulxBuilds /
Nitro
Nitro is a substantial typed TypeScript SEO product prototype with authenticated analysis and rank-tracking flows, Prisma/Postgres persistence, Browserbase scraping, and a Next.js client, but it has no visible adoption, documentation, tests, CI, or license.
MehulxBuilds /
connair
A small, typed Next.js and Actix-web paste-sharing application with SQLite persistence and a usable create/list/view/delete flow, but no documentation, tests, CI, license, or demonstrated adoption.
MehulxBuilds /
text-search
A compact, typed Go text-search CLI with stemming, stop-word filtering, line indexing, and multi-term intersection search, but minimal README content and no visible tests or automation.
MehulxBuilds /
rust-services
A compact Rust learning/examples repository spanning 11 sampled binaries and an Actix API, with typed code and a .gitignore but no README, tests, CI, license, or evidence of adoption.
MehulxBuilds /
agentic-stuff
An early-stage notebook collection exploring LangChain agents, iterative response evaluation, and PDF-to-Neo4j GraphRAG, with pinned dependencies and Docker infrastructure but no documented or tested delivery.
MehulxBuilds /
grpc-basics
A small Go gRPC learning project demonstrating unary, server-streaming, client-streaming, and bidirectional RPCs, with separated client/server handlers but no documentation, tests, CI, or license.
MehulxBuilds /
go-graphql-api
Documented, typed Go GraphQL Todo API with gqlgen and MongoDB integration, but it is a same-day, zero-star repository with no tests, CI, license, or demonstrated external adoption.
MehulxBuilds /
go-scraper
A one-commit Go scraper prototype with a small concurrent crawler in cmd/scrape/main.go, but no documentation, tests, CI, license, or repository configuration.
06 · Timeline
- Aug 29, 2024Joined GitHub
- Feb 28, 2026Created voxara — An Application made to destroy ElevenLabs
- Mar 27, 2026Created Elowen — Elowen is a chill AI chat app that lives right inside Telegram. No switching tabs, no extra apps — just message the bot and get smart replies instantly. Built with a Tuborepo, it’
- Jun 2, 2026Created RSSFeed
- Jul 10, 2026Created Nitro
- Aug 2, 2026Created go-scraper
- Aug 2, 2026Created rust-services
- Aug 6, 2026Created connair
- Aug 8, 2026Created agentic-stuff
- Aug 8, 2026Created go-graphql-api
- Aug 9, 2026Created grpc-basics
- Aug 12, 2026Created dora — An embedded key-value database engine for Go.
- Aug 25, 2026Created text-search
- Sep 1, 2026Most recent push to voxara
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.