Pranav Gujarathi

Pranav Gujarathi

Generative AI & ML Engineer. Building multi-agent AI for digital health at Twin Health. MS Data Science, Indiana University · IIT Roorkee.

About me

Hi, I’m Pranav

I’m a Generative AI & ML Engineer based in Austin, TX — and I genuinely believe we’re living through the most exciting moment in the history of computing. I’ve spent the last 8+ years turning that belief into production systems that actually move needles.

Right now, I work as a Senior AI Engineer at Twin Health, where I architect multi-agent AI platforms for personalized digital health. The work hits close to home in the most literal sense: the AI systems I build — ones that ingest real-time wearable data, interpret lab results, and deliver hyper-personalized diet and nutrition guidance — have contributed to a 37% Type 2 diabetes reversal rate across our patient population. That’s not a dashboard metric. That’s people reclaiming their health.

Before Twin Health, I spent over a year at Cigna Health building Generative AI into the Prior Authorization process — a notoriously slow, paper-heavy corner of healthcare that desperately needed agentic intelligence. And before that, I spent two years at Walmart Global Tech, where I built a voice/text-driven AI tool that saved 130,000 associate work hours every single week, and a RAG-based system that cut attribute-extraction costs by 80% with 24× faster turnaround. Large-scale impact is something I’ve come to expect from this work.

My path here wasn’t strictly linear. I started as an Electrical Engineering undergraduate at IIT Roorkee, pivoted into data science at Indiana University – Bloomington (where I earned a Master’s and the Luddy Outstanding Research Award), and along the way published research at NeurIPS (Best Paper) and ACM COMPASS. More recently, I completed an MBA in Entrepreneurship from University of the Cumberlands — because building great AI isn’t just an engineering challenge, it’s a product and people challenge too.

Publications

I’ve had the privilege of presenting research at some of AI’s top venues.

  • “Controlled-rearing studies of newborn chicks and deep neural networks”
    Shared Visual Representations in Human & Machine Intelligence workshop, NeurIPS 2021 — Best Paper Award
    arXiv · Workshop

  • “Using Causality to Mine Sjögren’s Syndrome related Factors from Medical Literature”
    ACM SIGCAS/SIGCHI Conference on Computing and Sustainable Societies (COMPASS)
    DOI

  • Luddy Outstanding Research Award, Indiana University – May 2021

Experience

Senior AI Engineer — Twin Health

Austin, TX · January 2026 – Present

  • Architecting a production-grade multi-agent AI platform for personalized digital health — specialized agents deliver end-to-end diet, nutrition, and recipe recommendations, contributing to 37% Type 2 diabetes reversal across the patient population.
  • Engineering a real-time continuous health monitoring agent that ingests lab results and wearable biosensor streams to surface personalized clinical insights and enable proactive patient intervention.

Senior AI Engineer — Cigna Health

Dallas, TX · September 2024 – December 2025

  • Spearheaded a Generative AI-augmented Prior Authorization platform built on agentic AI systems, robust ML-Ops pipelines, and seamless clinical workflow integration.
  • Led cross-functional teams delivering compliant, high-reliability AI solutions across medical care workflows.

AI Engineer — Walmart Global Tech

Dallas, TX · October 2023 – August 2024

  • Built a voice/text-driven Generative AI tool for exploratory data analysis — eliminating legacy dashboards and saving 130,000 associate work hours per week.
  • Delivered an end-to-end RAG-based attribute extraction and competitor analysis system achieving 80% cost savings and 24× faster turnaround vs. manual methods.
  • Led ML engineering for containerized API deployments via Docker and Kubernetes.

Data Scientist — Walmart Global Tech

Dallas, TX · January 2022 – October 2023

  • Deployed a large-scale anomaly detection engine with real-time feedback, achieving > 70% capture rate.
  • Contributed to a novel causal-inference forecast model explaining > 80% of variance with < 5% global error, underpinning ~$1.6B in sales analysis.
  • Built and deployed REST API solutions on Azure with CI/CD best practices; ran Gen AI PoCs including automated competitor price mining and a text-based forecasting interface.

Data Science Associate — ZS Associates

Los Angeles, CA · June 2018 – December 2021

  • Implemented an NLP pipeline using deep learning and transformer models to extract domain-relevant inferences from news and publications — deployed as a cross-platform product with strong client reception.
  • Launched an ML-based marketing strategy solution using multivariate time-series models and linear optimization, projecting 60% improvement in target reach and ROI.

Research Engineer — Indiana University – Bloomington

Bloomington, IN · January 2020 – May 2021

  • Mind Lab (NSF-funded): Designed and implemented Computer Vision and Deep Reinforcement Learning pipelines for behavioral research under Prof. Justin Wood.
  • IUPUI Data Lab: Conducted NLP research from ideation to publication under Prof. Sunandan Chakraborty.
  • Kelley School of Business: Deployed an end-to-end MLOps pipeline — from PoC to GUI dashboard — using big-data libraries and cloud-based parallel computing.

Technical Skills

  • Programming — Python (Expert), C/C++, MySQL
  • ML & AI Frameworks — PyTorch, TensorFlow, LangChain, LlamaIndex, Hugging Face Transformers
  • Generative AI — Multi-agent systems, RAG pipelines, LLM-Ops, prompt engineering
  • DevOps & Deployment — Docker, Kubernetes, CI/CD, Git
  • Cloud Platforms — AWS, Microsoft Azure, GCP
  • Data Engineering — Spark, Databricks, SQL, time-series modeling, statistical analysis

Beyond the Work

A few things that keep me curious and grounded outside of building AI:

I’m genuinely excited about AI in legal work — the legal system is one of the most access-constrained domains in society, and intelligent automation has an enormous opportunity to level the playing field. Similarly, I think deeply about AI for social good and medical literacy: the same technology I use to build clinical tools could help millions of people understand their own health, navigate complex systems, and make better decisions.

On a lighter note — I’m a tennis player who enjoys the mental chess of the game as much as the physical workout. I’m also a dedicated barbecue enthusiast (low-and-slow is the only way), and I follow the EV space closely, fascinated by how electrification is reshaping mobility, energy, and urban design all at once.

Feel free to reach out — whether it’s about AI, career journeys, or the perfect brisket temperature. I love a good conversation.