About Me

    # About Me

    Associate Architect at Tekion, deeply passionate about transforming developer workflows through AI. Currently building production AI systems with LangChain/LangGraph, implementing RAG pipelines with Vector Databases, and architecting intelligent agents using Model Context Protocol (MCP). My work focuses on creating AI solutions that understand code context, development patterns, and deliver measurable impact.

    My technical journey has been shaped by a simple belief: good architecture makes hard problems feel easy. Whether it's designing microservices that scale, optimizing Docker builds, or implementing MQTT for real-time IoT communication, I focus on creating solutions that developers actually want to use.

    My Approach

    I've worked across the full stack - from React and TypeScript on the frontend to Golang & Node.js on the backend. But what really excites me now is the AI infrastructure layer: LangChain orchestration, RAG implementations, Vector Database optimization, and production observability with LangFuse. There's something deeply satisfying about watching AI agents handle complex workflows gracefully while maintaining reliability and performance.

    Recent focus areas include building context-aware AI agents with LangGraph, implementing production-grade RAG systems, leveraging Vector Databases for semantic search, and finding the right balance between automation and human judgment in AI-powered development tools.

    Professional Journey

    Over 9+ years, I've scaled teams from 3 to 35+ engineers while architecting solutions adopted by 250+ developers organization-wide. From building IoT platforms deployed across General Motors dealerships to achieving 50% build time reduction through Docker optimization, I've consistently focused on making engineering more enjoyable and efficient.

    Key achievements include: - Architecting production AI systems with LangChain/LangGraph and RAG pipelines - Core contributor to AI workflow framework using Model Context Protocol (MCP) - Reduced OEM deployment cycles from 6+ months to weeks through platform engineering - Built Connected Displays IoT solution handling millions of MQTT requests - Achieved 4x cost reduction through build-once, promote-everywhere CI/CD strategy

    Beyond Coding

    I believe the best code is the code you don't have to write, and the best systems are the ones that feel invisible to their users. Always interested in discussions about AI agents, RAG architectures, system design, developer tooling, or how we can make engineering more enjoyable with intelligent automation.

    Outside of work, I enjoy gaming on my PC or PS5—it's my way of relaxing and staying inspired by great user experiences. Feel free to reach out!

    Arkadyuti Sarkar - Professional photo of Associate Architect @ Tekion with expertise in React, TypeScript, AI, and full-stack development

    Arkadyuti Sarkar

    Associate Architect @ Tekion

    LangChain/LangGraph
    LangFuse
    React.js (9+ years)
    Next.js
    Golang
    Node.js (8+ years)
    Get in touch

    My Skills & Expertise

    AI & Emerging Tech

    LangChain/LangGraph
    LangFuse
    Model Context Protocol (MCP)
    RAG (Retrieval-Augmented Generation)
    Vector Databases
    Prompt Engineering

    Frontend Development

    React.js (9+ years)
    Next.js
    TypeScript
    Redux
    Tailwind CSS
    HTML5
    CSS3
    JavaScript (ES6+)
    Webpack

    Backend Development

    Golang
    Node.js (8+ years)
    RESTful APIs
    Microservices Architecture
    Python
    PHP

    Cloud & DevOps

    AWS (EC2, S3, Lambda, IoT Core)
    Docker
    CI/CD Pipeline Design
    Infrastructure as Code
    Jenkins

    Database & Messaging

    MongoDB
    MySQL
    Redis
    MQTT
    SQS

    IoT & Hardware

    IoT Architecture
    MQTT Protocol
    OTA Updates
    Android Development
    Remote Debugging

    Testing & Quality

    Jest
    ESLint
    Unit Testing
    Integration Testing
    Performance Testing

    Architecture & Leadership

    Component Libraries
    Design Systems
    Technical Leadership
    Team Scaling
    Mentoring
    Cross-functional Collaboration