Pratham Sharma profile photo 1 on Tal

Pratham Sharma

AI Engineering Lead | Generative AI, LLM & ML Engineer | Agentic AI, RAG, LangGraph, MCP | Healthcare AI & FHIR

Hyderabad

2+ years of work experience

ilumina Health
Jun 20266 mosPresent

Team Leading , System Design automation owner
ilumina Health
Dec 20256 mosMay 2026

Developed a MCP Server over FHIR data that lets clinical staff query patient records in natural language Built a gen ai pipeline to extract structured diagnoses from physician notes, reducing manual review time. Built and deployed automation healthcare systems for operations smoothness
INZINT
Feb 20241 yr 10 mosNov 2025

Noida, Uttar Pradesh, India

Pratham’s AI Stack

claude code
claude code
claude
claude
gemini
gemini
hugging face
hugging face
langchain
langchain
githubcopilot
githubcopilot

Education

Inderprastha Engineering College

Bachelor of Technology - BTech, Computer Science ( Artificial Intelligence and Machine Learning )

Inderprastha Engineering College

20204 yrs2024

Join Pratham and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Pratham Sharma on Tal

Headline

AI Engineering Lead | Generative AI, LLM & ML Engineer | Agentic AI, RAG, LangGraph, MCP | Healthcare AI & FHIR

About

Building agentic AI systems in production — not demos. Currently architecting a AI platform at Ilumina Health: MCP servers over clinical data, real-time GenAI agents, and the infrastructure that makes LLMs actually reliable in healthcare. I lead Claude Code adoption across an 8-engineer team through context engineering and prompt design, and have built: → FHIR-compliant clinical data pipelines processing hundreds of clinical records daily, in line with interoperability standards → An MCP server over FHIR data, enabling natural-language querying of clinical records for care teams → A WhatsApp GenAI chatbot with tool invocation and session-managed responses, handling 100–1,000+ conversations/day → Agentic web scraping + OCR pipelines that cut manual operational processing time by 50% Before this, I spent two years at INZINT building the pieces that make agentic AI useful in production: → A real-time voice AI agent (Twilio + STT/TTS + LangChain) for natural telephonic conversations → Multilingual RAG-based customer support chatbots that increased adoption by 15% → Self-correcting agentic scraping systems with automated lead outreach Stack: Python · LangChain · LangGraph · MCP · AI Agents · Claude Code · RAG · Vector DBs · AWS · MongoDB · Twilio · Whisper/Deepgram What I care about: agentic systems that don't fall apart under real-world messiness — dynamic websites, ambiguous voice input, clinical data that doesn't fit clean schemas. LangGraph and MCP are my daily tools, but the actual skill is knowing when an LLM call is the wrong answer to a problem. Open to conversations about agentic AI, healthcare AI infrastructure, and roles where I can build systems like this at scale.

Location

Hyderabad

AI stack

claude code

claude

gemini

hugging face

langchain

githubcopilot

Work experience

AI Engineering Lead at ilumina Health, Jun 2026 - Present · 6 mos, Team Leading , System Design automation owner

AI Engineer at ilumina Health, Dec 2025 - May 2026 · 6 mos, Developed a MCP Server over FHIR data that lets clinical staff query patient records in natural language Built a gen ai pipeline to extract structured diagnoses from physician notes, reducing manual review time. Built and deployed automation healthcare systems for operations smoothness

Associate Software Engineer ( AIML) at INZINT, Feb 2024 - Nov 2025 · 1 yr 10 mos, Noida, Uttar Pradesh, India

Education

Bachelor of Technology - BTech, Computer Science ( Artificial Intelligence and Machine Learning ), Inderprastha Engineering College

Pratham Sharma - AI Engineering Lead in Hyderabad | Tal