Note: the WakaTime card above only renders once WakaTime is connected to this GitHub account — remove it if you don't use WakaTime, see setup notes at the bottom.
GenAI / LLM Engineering
Full-Stack Development
Core Languages, ML & CV
Tooling & Platforms
Skill icons above pulse/animate on hover via skillicons.dev's built-in rendering — no extra config needed.
⚡ Proficiency Snapshot
🔥 These bars are self-rated, not auto-measured — adjust the numbers whenever your comfort level shifts.
Four agents — Planner, Researcher, Executor, Critic — run as an explicit LangGraph state machine over documents and live tools, rather than a loose prompt-chaining pipeline. The goal is a multi-agent system that behaves predictably like software: typed state transitions, inspectable execution, and graceful fallback instead of silent hallucination.
Highlights
- Explicit state-machine graph (not implicit agent chatter) for traceable execution
- Grounded document retrieval feeding the Researcher agent
- LiteLLM-backed model routing for provider fallback resilience
- PostgreSQL persistence via Alembic-managed migrations
Real-time speech recognition wired to LLM integration for hands-free system automation, reminders, and live info lookup. Built as a React/Vite frontend backed by a Node.js/Express/Socket.io server for live event streaming between voice input and assistant response.
Highlights
- Continuous speech-to-intent pipeline with low-latency response streaming
- Deployed with a split strategy — frontend on Netlify, backend on Railway
- Designed as a primary showcase project for full-stack + AI integration skill
Ingests industrial PDFs, P&IDs, scanned forms, and logs into a unified, natural-language-queryable knowledge graph — built to turn messy, heterogeneous industrial documentation into something an LLM can actually reason over.
Highlights
- Multi-format ingestion pipeline (scanned + digital documents)
- Vector retrieval layer over a structured knowledge graph
- Natural-language query interface over industrial data
A complete MERN hostel management system built from scratch, with role-based dashboards for Admin, Warden, and Student, and Socket.io live updates across the app.
Highlights
- Six modules: Room Allocation, Fee Management, Complaints, Notices, Visitor Logs, User Management
- JWT authentication with role-based access control
- React/Vite/Tailwind frontend, Node.js/Express/MongoDB/Socket.io backend
Classifies MRI scans as tumor-positive or normal via a CNN pipeline with curated train/validation splits — classical CV/ML work alongside the GenAI-heavy projects.
A wide range of MERN apps demonstrating REST API design and modern JavaScript across many use cases — the breadth counterpart to the two flagship deep-dive projects above.
🎬 Netflix Clone (Frontend) — React · Vite · Tailwind · TMDB API
Full frontend scaffold in place; actively building out the MovieCard, the full ContentRow carousel, and TitleModal components, plus wiring in real Firebase Auth to replace the placeholder auth flow.
🧭 Conductor Hardening — LangGraph · LiteLLM · PostgreSQL
Taking Conductor from "working" to "production-styled": end-to-end verification of the four-agent flow, live LiteLLM fallback wiring, and running Alembic migrations against a live PostgreSQL instance rather than local dev.
⚙️ Animated snake activates once the daily GitHub Action is added — see setup note at the bottom.
Second activity graph, different theme/palette — remove this one if a single graph feels like enough motion on the page.
🏗️ Shipped 6 portfolio-ready repos, spanning GenAI orchestration, RAG, MERN, and classical CV 🧭 Conductor — went from prototype to an explicit 4-agent LangGraph state machine 🎤 Nexus AI Assistant — real-time voice pipeline deployed across two hosts (Netlify + Railway) 📚 KnowledgeBrainAPI — unified ingestion for scanned + digital industrial documents 🔥 12+ MERN builds in the general portfolio, on top of the two flagship deep-dives 🎯 Currently pushing Conductor and the Netflix clone toward production polish
🎉 This list is manually maintained — add a line here whenever you ship something worth bragging about.
| Area | What I'm sharpening |
|---|---|
| 🕸️ Multi-Agent Systems | Deeper LangGraph orchestration patterns beyond simple chains |
| 🏗️ System Design | Fundamentals for designing agent state machines that scale |
| 🔎 RAG Infrastructure | Retrieval quality, chunking strategy, and hybrid search over FAISS/ChromaDB |
| 🌐 Full-Stack Delivery | Shipping polished, real-time MERN features end-to-end |
Requires a small free deployment of novatorem linked to your Spotify account — see setup note at the bottom. Delete this section if you'd rather not run that.



















