01 · Roasts
Zero-star systems lab
The paged-KV engine claims 4.4x concurrency and has a live demo, but its 0 stars say the benchmark audience is still mostly you.
CI selective
agentgate has Docker-backed CI, gitleaks, and acceptance tests; several other substantial repos still treat CI as optional DLC.
Sprint-powered portfolio
meeting-intelligence-platform logged 30 sampled commits in six days, while web-automation-agent was born and pushed on the same day.
Horizontal builder
137 recent commit samples across repositories is serious output volume; the missing piece is turning more of it into adopted, maintained products.
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% weight65C
- Quality20% weight75B
- Depth15% weight58D
- Breadth10% weight95S
- Community10% weight25F
03 · Stats
365-day commit heatmap
90 active days
Language distribution
- Jupyter Notebook62%
- TypeScript12%
- Go8%
- Python8%
- JavaScript3%
- Java3%
- Other4%
04 · Numbers
Owned repos
non-fork
60
Commits
last 12 months
296
Followers
6
Joined GitHub
Aug 2024
05 · Top repos
dibyo10 /
agentgate
AgentGate is a documented, typed TypeScript measurement service with CLI and Fastify web/API surfaces, durable BullMQ/Postgres/Redis workers, SSRF and domain-policy controls, cost accounting, and extensive lifecycle/acceptance coverage; adoption is not yet demonstrated.
dibyo10 /
paged-kv-cache-llm-inference-engine
A technically ambitious Go/Python paged-KV inference server with a live demo, real gRPC control-plane/worker split, strong allocator and model-correctness tests, but no visible adoption, CI, or license yet.
dibyo10 /
type-ahead-system-for-scale
A documented, typed Go typeahead service with trie ranking, logical distributed caching, Postgres durability, trending scores, and batched writes; implementation is structured but lacks tests, CI, and adoption evidence.
dibyo10 /
v0-modern-portfolio-design
A deployed, polished TypeScript/Next.js portfolio with modular sections, Framer Motion effects, canvas particle systems, and a Three.js gradient-descent visualization; adoption is unproven and engineering validation is limited.
dibyo10 /
meeting-intelligence-platform
A substantial but early-stage AI meeting product: FastAPI/SQLite ingestion, Whisper and pyannote processing, Gemini extraction, ChromaDB RAG, analytics, and a React UI, with strong design documentation but no tests, CI, license, or demonstrated adoption.
dibyo10 /
Distributed-Cache
A documented Go cache prototype with tested O(1) LRU, mutex/sharded cache variants, HTTP node endpoints, and a consistent-hash package, but Phase 2 routing remains unimplemented and adoption is currently absent.
dibyo10 /
web-automation-agent
A documented JavaScript Google ADK + Gemini + Playwright browser-automation agent with modular services, DOM/vision targeting, and smoke/wiring scripts, but it is a same-day, zero-star project without CI or tests.
dibyo10 /
devops-homework
A documented DevOps homework repository covering seven Linux, shell, networking, Git, and Docker exercises, with runnable Compose examples and a static validation script, but no demonstrated adoption or automated CI/tests.
dibyo10 /
dibyo10
A personal GitHub profile configuration repository with a README full of stack badges and one scheduled contribution-snake workflow; it has presentation value but no substantive software product or tested implementation.
06 · Timeline
- Aug 20, 2024Joined GitHub
- Jun 11, 2025Created dibyo10 — Config files for my GitHub profile.
- Jan 1, 2026Created v0-modern-portfolio-design
- Jun 18, 2026Created meeting-intelligence-platform
- Jun 21, 2026Created type-ahead-system-for-scale
- Jun 24, 2026Created web-automation-agent
- Jul 3, 2026Created Distributed-Cache
- Sep 1, 2026Created paged-kv-cache-llm-inference-engine — Paged KV cache inference server. Go control plane, Python worker. 97% pool utilization vs 22% baseline, 4.4x concurrency, output bit-identical to HuggingFace.
- Sep 2, 2026Created devops-homework — DevOps homework: Linux, shell scripting, networking, Git and Docker exercises with runtime evidence
- Sep 2, 2026Created agentgate
- Sep 3, 2026Most recent push to dibyo10
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.