LightSpun
AI Researcher – Healthcare
- Built healthcare GenAI and agentic workflows with grounding and safety constraints.
- Applied RAG / LLM patterns to clinical and operations knowledge use cases.
I design AI systems that ship.
The interesting part of AI is not the demo — it is the architecture that makes it useful, safe, and durable in production.
8+ years building applied AI across telecom, finance, and healthcare GenAI. I design LLM systems, RAG / GraphRAG pipelines, multi-agent workflows, and the evals, tooling, and cloud platforms that keep them running after launch.
M.Tech in Data Science at IIT Hyderabad — graduated — with hands-on work across Vertex AI, LangGraph, MCP, and production LLMOps. Open to AI Architect / GenAI roles worldwide and selective consulting.
- GenAI · Agents · RAG
- Healthcare · Telecom · Finance
- GCP · Vertex · LLMOps
- M.Tech · IIT Hyderabad
Open to AI Architect roles · Worldwide
Technical expertise
What I use to design and ship production AI — grounded in GenAI agents, classic ML, and cloud platforms.
Core skill groups
Agents & protocols
Retrieval & models
Evals & ops
Languages & APIs
ML · DL
Cloud · Platform
Architecture practice
Selected work
Flagship engagements that show how I take GenAI and ML from idea to production.
Healthcare · GenAI
Healthcare GenAI & agentic workflows
Production-oriented Generative AI for clinical and operations use cases at LightSpun.
Outcome: Faster knowledge access and safer LLM-assisted workflows in regulated settings.
Telecom · Network AINetwork intelligence at Airtel scale
Lead Data Science for network analytics — models that inform planning and performance.
Outcome: Decision systems grounded in large-scale telco data, not demos.
Finance · ML SystemsFinance ML systems
Senior Data Science for finance AI — robust models and delivery under enterprise constraints.
Outcome: Reliable ML delivery aligned with risk and compliance expectations.
Healthcare · Document AIDocument AI + grounded RAG
Turn messy healthcare documents into retrieval-ready knowledge with PHI-aware grounding.
Outcome: Faster, cited answers from policies and forms — fewer hallucinated ops lookups.
Platform · AgentsAgentic AI with MCP & ADK
Multi-tool agent architecture on Vertex — MCP tools, ADK orchestration, production guardrails.
Outcome: Operable agents with traceable tool calls, not demo chatbots.
Public artifacts
Things hiring managers can click — code, writing, and longer-form notes.
How I help
Hands-on architecture and advisory for teams shipping GenAI, not slideware.
AI architecture
End-to-end system design for LLMs, retrieval, evaluation, and deployment paths that survive production.
RAG & LLM production
Retrieval pipelines, grounding, monitoring, and iteration loops so GenAI features stay useful after launch.
Advisory & enablement
Roadmaps, reviews, and pairing with your team across banking, retail, telecom, and healthcare contexts.
Professional journey
8+ years building AI solutions — from Ericsson (Feb 2018) through healthcare GenAI today. Full narrative on the resume and LinkedIn.
IRIS Software
Senior Data Scientist – Finance
- Delivered finance-focused ML under enterprise risk and compliance constraints.
- Partnered with stakeholders on validation, ownership, and production readiness.
Bharti Airtel
Lead Data Science – Network
- Led network analytics ML for planning and performance at telco scale.
- Owned delivery from problem framing through operational handoff.
L&T Technology Services
Engineer – Machine Learning
- Machine learning engineering and applied modeling for industrial / product contexts.
- Worked with Intel frameworks and production-minded ML pipelines.
Ericsson
Data Scientist
- Career start: February 2018 — applied data science in telecom domains.
- Time series forecasting and classical ML (ARIMA, Naive Bayes, and related methods).
Writing
Notes on GenAI architecture and shipping LLMs — also on Medium.
Education & certifications
Formal degrees plus continuous learning across AI engineering, GenAI, cloud ML, and IoT.
M.Tech: Data Science 8.0/10
Indian Institute of Technology, Hyderabad
- Data science, machine learning, and applied GenAI research focus
B.Tech: ECE 8.5/10
JIS College of Engineering (WBUT)
- Top 1% in the department · Best undergraduate researcher
Generative AI with Large Language Models
Specialization · AWS / Coursera track
- LLM fundamentals, fine-tuning, and production GenAI patterns
Master's Program: AI Engineering
Simplilearn
- ML, deep learning, NLP, Spark — multi-subject industry curriculum
AWS Machine Learning Nanodegree
Udacity
- Cloud ML workflows, deployment, and evaluation on AWS
Internet of Things (XEE100-017)
Stanford University
- Embedded systems, sensors, networking, and IoT applications
Let's talk
Open to AI Architect / GenAI roles and selective consulting. Best first step: email with role, timeline, and problem space.