Hi, I am Shreyas

ML Engineer | AI Agents & LLMs | Building Intelligent Solutions | Software meets ML

Building innovative AI solutions and scalable architectures that drive business transformation.

Scroll to explore

Things I Love

Hiking Bay Area

Love exploring beautiful trails and scenic spots around the Bay Area. There's nothing like a good hike to clear the mind and connect with nature.

Daily Gym Sessions

Consistency is key! I hit the gym daily because I believe in maintaining both physical and mental strength. Discipline in fitness translates to discipline in work.

Music Lover

Music fuels my creativity and keeps me motivated throughout the day. Whether it's coding or working out, the right playlist makes everything better.

Formula 1 Fan

The precision, strategy, and cutting-edge technology in F1 fascinate me. It's the perfect blend of human skill and engineering excellence - much like ML engineering!

A Small Selection of Recent Projects

Fast F1 Streaming Analytics

This project streams F1 data through Kafka and processes it with Spark, using HyperLogLog and Bloom Filters to handle high-volume analytics efficiently. The goal is to build a real-time, scalable pipeline for fast insights on racing events.

View Repository

Project Screenshot

Task Management App

A collaborative task management platform with real-time updates, team collaboration features, and advanced analytics dashboard.

Check Live Site

ClaimsGuard – AI Policy/Claims Assistant

An autonomous agent that tracks payer/CMS policy updates, stores them in a vector DB, and flags impacted claims. Built with Python, FastAPI, ChromaDB, LLMs, and Playwright.

Checkout

Experience

Jan 2022 - Dec 2024

Senior Data Analyst

EXL Health

Delivered 89% accuracy in bill-type classification by implementing optimized K-means clustering models with PySpark.
Performed exploratory data analysis (EDA) to identify key drivers of overpayments in healthcare claims. Developed a predictive model to flag high-risk claims, enabling proactive audit selection and reducing financial leakage.
Deployed Hive and PySpark pipelines to run MoM overpayment audits, uncovering $300K in contract-level discrepancies.
Managed analytics for a $150M+ revenue generating focus group by developing a provider level framework that enhanced clinical finding accuracy and audit precision.
Automated concept development using Selenium and Python, and built Databricks BI dashboards to monitor audit performance and data spend, saving 10+ hours/week and enabling data-driven claim selection decisions.

Jun 2021 - Dec 2021

Software Developer Intern

Buddi.ai

Enhanced Buddi.ai's proprietary Medical Language Processing (MLP) engine structured & unstructured EMR notes, PDFs, and HL7 data into a Clinical Contextual Graph, achieving over 95% accuracy in clinical information extraction.
Built RESTful APIs with FastAPI for backend integration, improving processing time by 25%.
Designed an HL7 parser to extract and normalize structured healthcare data, increasing system interoperability.
Wrote unit test cases to ensure robustness and accuracy of the data extraction pipeline, improving reliability and reducing debugging time during integration.