Freelancers using GitLab in Pallavaram (Contonment)

Freelancers using GitLab in Pallavaram (Contonment)Freelancers using GitLab in Pallavaram (Contonment)

Projects People

Pallavaram (Contonment), India

Backend & systems dev — Rust, Go, Python, databases

New to Contra

Backend & systems dev — Rust, Go, Python, databases

Cover image for A browser-based tool that turns

A browser-based tool that turns any GitHub repo into an instant, visual "intelligence report." You paste a GitHub URL and get a full dashboard back — no sign-up, no auth, no friction. It's a developer-tool / analytics product. Core features Health Score (0–100) — production-readiness rating across 7 quality dimensions, with letter grades Language pie chart — interactive breakdown of language composition Commit heatmap — GitHub-style 52-week activity grid Contributor rankings — top 10 contributors with avatars + metrics Smart dependency detection — parses package.json, requirements.txt, Cargo.toml, go.mod, pom.xml, Gemfile File tree browser — collapsible directory viewer with file-type icons README rendering — full GitHub-flavored Markdown with syntax highlighting Share cards — export a PNG summary for social media Tech stack LayerTechFrontendReact 18, Vite, Tailwind CSS, Recharts, react-markdownBackend (optional)Python 3.12, FastAPI, httpx, file-based caching, slowapi rate limitingDeployGitHub Pages (frontend), Railway (backend, optional) How it works The frontend runs entirely in the browser and talks directly to GitHub's REST API — that's why the live demo needs no setup or login (limited to ~60 requests/hour unauthenticated). The FastAPI backend is optional, and only exists to add caching and rate-limit relief when you self-host. Run it locally Or docker compose up --build.

0

15

Cover image for AI Agent Skills Workflow Framework Development

AI Agent Skills Workflow Framework Development

0

1

Cover image for Most RAG systems have two

Most RAG systems have two problems that nobody talks about enough. The first is semantic drift — your retriever pulls in chunks that look relevant (high cosine score) but don't actually answer the question causally. Ask "Why did Lehman Brothers collapse?" and you'll get back chunks about the 2008 housing crisis — same vocabulary, but those are the consequences, not the cause. Cosine similarity can't tell the difference. The second is context poisoning — even if each individual chunk is okay, a window full of semi-relevant chunks confuses the LLM. It attends to all of them, averages them out, and hallucinates. VORTEXRAG fixes both. It's a 7-layer pipeline I built specifically around these two failure modes. Each layer has a specific job:

0

12

Cover image for Development of datamend Python Library

Development of datamend Python Library

1

3