Experience

  1. October 2025 – August 2026

    Bangalore, Karnataka (Remote)

    Samsung PRISM R&D Institute

    R&D Intern

    • Built Guide Weave, an AI tool that turns product manuals into step-by-step visual guides, using a multimodal RAG pipeline with ChromaDB to pull matching text and images from the manuals.
    • Ran the LLM locally so no data left the machine, and wrote benchmarks to raise retrieval accuracy and cut response time.
    • Multimodal RAG
    • ChromaDB
    • Local LLM
    • Python
    • Benchmarking
  2. June 2026 – August 2026

    Bangalore, Karnataka (Remote)

    Aparsoft Private Limited

    Software Engineering Intern

    • 12-week internship in full-stack and agentic AI work on an educational AI chatbot platform (Django REST Framework, Next.js, LangChain/LangGraph).
    • Found and fixed bugs in the Next.js frontend of a RAG system backed by PostgreSQL (pgvector) and Redis.
    • Next.js
    • Django REST
    • LangChain
    • PostgreSQL (pgvector)
    • Redis

Open source

  1. 2026

    GitHub (Remote)

    Activity Frames

    Open Source Contributornossa-y/activity-frames

    • Contributed to activity-frames (the engine behind Nocta), which turns local screen activity into structured frames that AI agents can read.
    • Worked on the Python and Model Context Protocol (MCP) code that compiles sessions, so agents can replay a workflow and build context without spending tokens.
    • Python
    • MCP
    • SQLite
    • React
    • TypeScript
    • GitHub