Hi, I'm

Manas Choudhary

Building AI-powered solutions

I'm an AI engineer specializing in full-stack development. I build intelligent applications and contribute to open source, creating impactful solutions that bridge technology and innovation.

LinkedIn Journey
OpenPeeps character illustration
// NEXT: PROJECTS
Strategic Code Companion
TypeScriptAI AgentsASTBabelD3.js

Strategic Code Companion

AI-Powered Code Analysis Platform

VS Code extension and CLI tool providing semantic-aware multi-agent AI analysis for comprehensive code review. Features 4 specialized agents (Refactor, Architect, Librarian, Tutor) with intelligent chunk selection using multi-dimensional scoring. Analyzes business flows, architectural patterns, and quality hotspots with D3.js visualization and PDF reporting.

AI-Powered Figma-to-React Converter
TypeScriptMCPLLMFigma API

AI-Powered Figma-to-React Converter

Design-to-Code Automation

Enterprise-grade design-to-code automation system using MCP (Model Context Protocol) for AI agent orchestration. Engineered design extraction pipeline with LLM-based code generation achieving 70-80% time savings with TypeScript safety and WCAG accessibility compliance.

RagAgent
PythonRAGNeo4jAI AgentsFastAPI

RagAgent

Multi-Agentic Hybrid RAG System

Intelligent insurance policy analysis system combining 6 specialized AI agents with Neo4j knowledge graph. Evaluates coverage eligibility, calculates payouts, and provides explainable decisions with cited policy clauses. FastAPI backend with hybrid RAG approach grounding LLM responses in actual policy relationships for transparent, accurate claim processing.

AI Career Platform
PythonReactFastAPIGeminiLangChain

AI Career Platform

Intelligent Career Matching

Full-stack career development platform featuring 6 specialized agents (CareerAnalyst, MentorBot, SkillsAssessor, LearningOrchestrator, OpportunityScout, ProgressMonitor). Implements Big Five & RIASEC assessments, career fit scoring with reasoning, context-aware mentor chat with sentiment analysis, and analytics dashboard for market trends. React 19 + TypeScript frontend with FastAPI backend and SQLite database.

TripPlanner
PythonLangChainGoogle PlacesFastAPIGemini

TripPlanner

AI-Powered Travel Assistant

LangChain ReAct agent powered by Gemini 1.5 Flash for personalized travel planning. Integrates Google Places data, review scraping, and preference-based filtering (city, time, place type, food, budget). Generates day-wise itineraries with 4.0+ rated venues using multi-source data aggregation. FastAPI backend with Swagger UI and extensible tool architecture for real API integrations.

Gmail RAG Assistant
PythonRAGChromaDBLangChainGemini

Gmail RAG Assistant

Intelligent Email Assistant

RAG system enabling natural language queries over Gmail inbox. OAuth 2.0 authentication with Gmail API for secure email access. ChromaDB vector database for semantic search, LangChain for RAG pipeline orchestration, and Gemini 1.5 Flash for intelligent responses. Gradio web interface with 2D vector space visualization. Privacy-focused with local processing and 1-3 second query response times.

Technologies I Master
Tech iconTech iconTech iconTech iconTech iconTech iconTech iconTech iconTech icon

TECH UNIVERSE

PythonPython
JavaScriptJavaScript
TypeScriptTypeScript
C++C++
HTML5HTML5
CSS3CSS3
ReactReact
Next.jsNext.js
TailwindTailwind
Three.jsThree.js
FramerFramer
Node.jsNode.js
ExpressExpress
FastAPIFastAPI
FlaskFlask
GraphQLGraphQL
TensorFlowTensorFlow
KerasKeras
LangChainLangChain
CrewAICrewAI
RAGRAG
LLMLLM
LoRALoRA
MCPMCP
MongoDBMongoDB
FirebaseFirebase
Neo4jNeo4j
RedisRedis
AWSAWS
GCPGCP
DockerDocker
GitGit
GitHubGitHub
VercelVercel
React NativeReact Native
ExpoExpo

PAST INTERNSHIPS

Ever saw a 3rd sem guy getting 5 internships? Check it here 👇

Atlan logo

Associate Software Engineer

Atlan

Oct 2025 - Present | Singapore · Remote

Working as an AI Engineering Intern at Atlan, building agents that can think about code. Designing Code Quality Intelligence Agents that analyze and enhance developer output. Experimenting with MCP (Multi-Agent Collaboration Protocol) to enable autonomous AI agents to work together. Bridging machine learning, reasoning, and real-world engineering to make development smarter and faster.

Skills: AI Agents, Multi-Agent AI Systems, MCP

OPEN SOURCE

Loading GitHub stats...