I'm a Full-Stack AI Engineer with 3.5+ years building scalable web applications, now specializing in production AI systems. I design and ship AI-powered enterprise knowledge platforms for European clients — RAG pipelines, LLM orchestration with LangChain & LangGraph, and multi-agent workflows on a NestJS + Angular stack.
My foundation is the MERN stack and modern TypeScript ecosystems, evolved through production experience into LLM-driven architectures: vector search, agentic orchestration, and event-driven microservices. I write clean, maintainable code that solves real problems end to end.
What I bring to teams:
- Production AI: RAG pipelines, LangGraph multi-agent orchestration, vector/semantic search (Weaviate)
- Full-stack ownership from database design to Kubernetes deployment
- Modern tooling across NestJS, Angular, Next.js, Docker, Azure & AWS
- A pragmatic approach that balances speed with code quality
I'm actively seeking Full-Stack / AI Engineer roles where I can build meaningful AI-powered products and grow with a collaborative team.
📍 Based in Thiruvananthapuram, Kerala, India | 🌐 Open to remote opportunities
- AI Industrial Knowledge Platform — Building an AI-powered enterprise platform (RAG + multi-agent) for a European client, on Angular, Vue.js & NestJS
- Production RAG pipelines — Document ingestion, hierarchical chunking, vector embeddings & Weaviate semantic search across multiple tenants
- LangGraph multi-agent workflows — Decorator-based plugin system for tool discovery and LLM-as-router sub-agent orchestration
Frameworks & Orchestration — LangChain · LangGraph · Multi-Agent Orchestration · Agentic AI · Model Context Protocol (MCP) Retrieval & Search — RAG Pipelines · Vector Embeddings · Semantic Search · Weaviate (Vector DB) Models & Tooling — OpenAI API · Anthropic Claude API · Prompt Engineering · SSE Streaming · Langfuse (Observability)
Full-Stack AI Engineer — German AI company (industrial software) · Dec 2024 – Present Building an AI-powered industrial knowledge platform — production RAG pipelines, LangGraph multi-agent workflows, and multi-tenant auth on Angular, Vue.js & NestJS.
MERN Developer — Freelance / Self-Employed · Nov 2022 – Nov 2024 Shipped e-commerce platforms and event-driven microservices (Docker, Kubernetes, RabbitMQ) on AWS; cut API latency ~60% with Redis caching and Bull job processing.
Multi-agent service that tailors a resume to a job description without fabricating experience — a Google ADK pipeline (parallel + sequential agents, schema-enforced outputs) behind a FastAPI service with a PDF-upload web UI.
Python · Google ADK · Multi-Agent · FastAPI · Gemini · Pydantic
Multi-agent system that turns a single question into a researched report — four cooperating agents (Supervisor, Researcher, Writer, Reviewer) as nodes in a stateful StateGraph, with dynamic LLM-based routing, an iterative Writer–Reviewer refinement loop, and a provider-agnostic LLM layer (Groq, Gemini, OpenAI, Anthropic).
Python · LangGraph · LangChain · Pydantic v2 · Docker
Production-style RAG chatbot grounded in user PDFs (including scanned/image-only PDFs via Tesseract OCR), with visual, verifiable citations mapping each retrieved chunk back to bounding boxes on the source page. Semantic retrieval via Sentence-Transformers + FAISS with a TF-IDF fallback.
Python · LangChain · FAISS · Tesseract OCR · PyMuPDF · Streamlit
📊 This Week I Spent My Time On
🕑︎ Time Zone: Asia/Kolkata
💬 Programming Languages:
Markdown 2 hrs 41 mins ████████████████████░░░░░ 81.39 %
Lua 26 mins ███░░░░░░░░░░░░░░░░░░░░░░ 13.12 %
Other 9 mins █░░░░░░░░░░░░░░░░░░░░░░░░ 04.90 %
Python 0 secs ░░░░░░░░░░░░░░░░░░░░░░░░░ 00.47 %
Text 0 secs ░░░░░░░░░░░░░░░░░░░░░░░░░ 00.09 %
🔥 Editors:
Claude Code 3 hrs 7 mins ████████████████████████░ 94.43 %
VS Code 11 mins █░░░░░░░░░░░░░░░░░░░░░░░░ 05.57 %
🤖 AI Coding This Week
⏱ AI Coding Time: 3 hrs 17 mins (99.76%)
✍️ 1,263 lines written by AI, 0 lines written by hand (100.0% AI-written)
🔤 61,053,273 Input Tokens, 342,883 Output Tokens
💵 $341.83 Estimated AI Cost This Week
🧠 11 AI Sessions, 34 AI Prompts
Opus 1,174 lines ███████████████████░░░░░░ 76.33 %
Fable 364 lines ██████░░░░░░░░░░░░░░░░░░░ 23.67 %
Claude-Code 0 lines ░░░░░░░░░░░░░░░░░░░░░░░░░ 00.00 %
🔎 AI Coding Insights:
🤖 AI-Driven — 100.0% of written lines came from AI
📚 Verbose Prompter — average 10,416 characters per prompt
🔁 Iterative Prompter — average 3 prompts per session
🚀 High AI Trust — 0.0% of changed lines were hand-edited
Last Updated on 11/08/2026 19:35:45 UTC