Fullstack · Agentic AI · LLMs

Yash Barhate

I'm Yash Barhate — a fullstack developer working across agentic AI, LLMs, and programmer-first tooling. Final-year, building products in the meantime, and open to roles & freelance work.

Yash Barhate
Scroll

Core Capabilities

What I Build

01 Agentic Systems

Multi-Agent Architectures

Building autonomous, tool-using AI agents that plan, reason, and execute complex multi-step workflows — with state persistence, fallback loops, and deterministic execution.

02 Full-Stack Web

End-to-End Applications

Crafting complete production applications — from database schemas and fast API endpoints down to responsive, high-performance user interfaces.

03 AI Infrastructure

RAG & LLM Integration

Connecting LLMs to enterprise data using hybrid retrieval, vector search, function calling, and structured outputs for reliable software systems.

Tech Stack

Stack I use to ship

🤖

AI & Agentic Systems

Autonomous agents, LLM apps & RAG pipelines

  • Python
  • LangGraph
  • LangChain
  • CrewAI
  • MCP Protocol
  • RAG Pipelines
  • Vector DBs (Chroma)
  • LLM APIs (Claude, OpenAI)
  • Agentic Tooling
🌐

Full-Stack Engineering

Modern web frontends, backend APIs & WebSockets

  • TypeScript
  • JavaScript
  • React.js
  • Node.js
  • FastAPI
  • Express.js
  • Tailwind CSS
  • WebSockets
  • Vite
☁️

Cloud & DevOps

Containerization, CI/CD pipelines & cloud hosting

  • Docker
  • AWS
  • GitHub Actions
  • Nginx
  • Vercel
  • Render
  • Git & GitHub
📊

Databases & Data

Relational & NoSQL databases, analytics & automation

  • PostgreSQL
  • MongoDB
  • Supabase
  • n8n Workflows
  • Pandas & NumPy
  • Scikit-learn
  • Twilio API

Selected Work

Things I've
shipped

A collection of agentic AI platforms, fullstack web products, and developer tooling built with production engineering standards.

01

Agentflow AI — Visual AI Workflow Engine

A fullstack visual AI workflow automation platform executed via a 5-agent orchestration engine for DAG planning, LLM execution, grounded validation, and fault recovery.

Next.jsNode.jsExpressMongoDBSocket.IOOpenRouterGemini
⚡ Real-time execution monitoring with sub-10ms UI updates & 5-agent fallback recovery View Case Study ↗
Agentflow AI Visual Workflow Canvas
02

Tripset AI Travel Platform

An agentic travel platform that plans multi-step itineraries, queries live inventory, and isolates session states with MongoDB persistence.

ReactPythonLangGraphFastAPIMongoDBExpress
⚡ 45% latency reduction via thread session isolation & agent state caching View Case Study ↗
Tripset AI Planner Dashboard
03

WebGenix AI Site Builder

Prompt-driven web application builder that generates fullstack component architectures with live rendering and instant sandbox previews.

TypeScriptReactNode.jsClaude APIDockerVite
🚀 Generates production-ready component codebases in < 5 seconds View Case Study ↗
WebGenix AI Web Generator Interface
04

MCP Learning Path Engine

Model Context Protocol engine that analyzes developer skill graphs, parses repository dependencies, and builds targeted learning paths.

PythonLangChainMCP ProtocolFastAPIChromaDB
🎯 Parsed 100+ skill graphs into deterministic learning nodes View Case Study ↗
MCP Path Engine Workspace

Final Year · Open to Roles & Freelance

Let's Connect

Whether you're hiring for a full-time role, internship, or freelance project, I'd love to hear about it.

Or, skip straight to Resume