Skip to content

Jay Gadhiya

Full-Stack AI Engineer | Ahmedabad, India

hello@jaygadhiya.com | linkedin.com/in/jaygadhiya-ai | jaygadhiya.com

Printed from jaygadhiya.com

Jay Gadhiya
  • All work
  • Contact

All work

Four systems I built at Vibgyor Technologies, and one I built for myself.

  1. Vibgyor TechnologiesMulti-provider LLM platformSwitch between OpenAI, Gemini and AWS Bedrock with one environment variable and zero code changes.
    • Python
    • OpenAI
    • Google Gemini
    • AWS Bedrock Converse API
  2. Vibgyor TechnologiesAI exam grading engineGrades answer sheets for 3+ exam boards and replaces hours of manual marking per batch.
    • Python
    • FastAPI
    • PDF parsing
    • LLM-based evaluation
  3. Vibgyor TechnologiesReal-time healthcare agentsTwo real-time clinical agents. The diagnostic agent went through 4 versions.
    • OpenAI Agents SDK
    • OpenAI Realtime API
    • WebSockets
    • RAG
    • Vector stores
  4. Vibgyor TechnologiesSpring Boot to FastAPIAverage API response time fell by about 50%.
    • Python
    • FastAPI
    • SQL schema design
  5. Personal project, 2026Yaadi, an offline-first notes appAn installable notes app that works offline and syncs when it can, on a near-zero-cost serverless backend.
    • React
    • TypeScript
    • IndexedDB
    • FastAPI
    • AWS Lambda

Jay Gadhiya, Full-Stack AI Engineer, Ahmedabad

  • hello@jaygadhiya.com
  • LinkedIn
  • GitHub
  • Résumé (PDF)
  • Markdown version

© 2026 · Last updated October 2026