Education Experience What I Do Projects Contact Resume ↗
Based in the US

Chinmayi
Hegde.

Data Scientist. Production ML, responsible AI, and LLM systems — with a growing focus on product analytics and user-facing impact.

Download Resume View Work
Education
M.S. Computer Science
San Jose State University
2022 – 2024 · San Jose, CA
B.E. Computer Science
PES University
2016 – 2020 · Bangalore, India
Experience

Where I've
Worked.

Jan 2025 – Present
Bentonville, AR
Walmart
Senior Data Scientist
Walmart · AI Governance
  • Built an agentic AI workflow using Python to help non-developers assess AI use case feasibility and risk — originally developed at an internal hackathon, then hardened for scale and under review for adoption across 1M+ Walmart associates.
  • Designed fairness testing frameworks across fraud detection, employment, and finance models using Fairlearn and SHAP; surfaced disparate impact and sent models back for remediation.
  • Evaluated 100+ ML models to inform product deployment decisions and improve reliability across use cases.
GPT-4 Fairlearn SHAP Python AI Governance
Nov 2024 – Jan 2025
Pleasanton, CA
NASA
Data Scientist
NASA · AI for Life in Space
Contract via Crowdplat
  • Built a deep learning model (Python, AWS) to classify rodent cells from spaceflight vs. ground control experiments using image and tabular data for NASA's AI4LS research program.
  • Implemented explainable AI (XAI) functionality using Streamlit to improve model transparency and interpretability for researchers.
Deep Learning XAI AWS Streamlit Python
Sep 2024 – Nov 2024
Santa Clara, CA
Chegg
Data Scientist II
Chegg
  • Built causal impact models to assess marketing campaign effectiveness across 11 social media platforms.
  • Oversaw 5 A/B tests with Optimizely on refund and payment pages, delivering insights to reduce churn.
  • Automated reporting for perks activation among subscribers using AWS Redshift, Python, and Tableau.
Causal Inference A/B Testing Optimizely AWS Redshift Tableau
Jun 2023 – Aug 2023
New York, NY (Remote)
WebMD
Software Development Intern
WebMD
  • Fine-tuned BERT models with PyTorch for sequence labeling of 26,000+ product names; used Llama API prompt engineering achieving 92–94% accuracy.
  • Built an ML pipeline (Python, BigQuery, Vertex AI) for time series forecasting, delivering insights to 300K+ healthcare professionals.
  • Built propensity and customer segmentation models to optimize outreach for the email marketing team.
BERT PyTorch GCP BigQuery Vertex AI NLP
Jan 2020 – Jun 2022
Bangalore, India
Merkle
Data Scientist
Merkle (Dentsu)
  • Automated anomaly detection on 2.5TB of web analytics data using time series forecasting — reduced detection lag from hours to minutes.
  • Delivered 7+ Tableau dashboards powered by time series forecasting and NLP text summarization for advertisement and marketing clients.
  • Designed Airflow data pipelines integrating unstructured data from 20+ systems and resolving data quality issues.
Anomaly Detection Time Series Airflow Tableau Python SQL
What I Do
Product Data Science

I find out why your metrics are moving — and what to do about it

I turn user behavior data into product decisions. From designing A/B tests that surface real signal to building models that explain churn, engagement, and retention — I work at the intersection of data and product thinking, not just data pipelines.

LLMs & Production ML

I take AI from prototype to something users actually rely on

I've led LLM selection for a live enterprise system, owned knowledge base architecture in production, and built agentic workflows end to end. I understand the gap between a model that works in a notebook and one that holds up at scale.

Responsible AI

I make sure your models don't fail the people they affect

I audit production ML for bias, disparate impact, and failure modes that hurt users and create risk. Using tools like SHAP and Fairlearn, I surface what models can't explain — and I've reviewed fraud and employment models and sent them back when they didn't meet the bar, before they shipped.

Projects

Things I've
Built.

04

NASA Rodent Cell Classifier

Deep learning model to classify rodent cells from spaceflight vs. ground control experiments for NASA's AI4LS research program. Includes XAI layer built with Streamlit for researcher interpretability.

Deep Learning XAI AWS
05

Schedulize — Email to Calendar

Automates conversion of itineraries from emails into Google Calendar events using NLP parsing. Solves a real scheduling problem end-to-end.

NLP Python Google Calendar API
06

Multilingual Image Description Assistant

Streamlit app combining Hugging Face image-to-text, LangChain translation, and text-to-speech to describe images in multiple languages with audio output.

LangChain Hugging Face Streamlit
07

Finance Analytics with Llama 2

Automated financial tool using Python for forecasting and prompt engineering with Llama 2 for transaction categorization. Interactive Tableau dashboards for spending trends and forecasts.

Llama 2 Python Tableau
08

Anomaly Detection at Scale

Processed 2.5TB of behavioral data at Merkle — reduced detection lag from hours to minutes using time series forecasting.

Anomaly Detection BigQuery Python
09

Customer Churn Prediction

ML pipeline comparing five algorithms across two datasets. Covers end-to-end feature engineering, model selection, and evaluation — with customer lifetime value analysis to segment high-value groups.

XGBoost scikit-learn Python
Publication

Vehicle Trajectory Prediction using Generative Adversarial Networks

Built a generative AI system to predict moving vehicle paths for collision avoidance using TensorFlow, YOLOv3, and GANs.

GANs YOLOv3 TensorFlow
Read Paper ↗
Publication

Autonomous Defense Device with IoT and Natural Language Understanding

Developed a safety device for self-defense with an AI voice assistant and automatic SOS location reporting via GPS and NLU.

IoT NLU GPS
Read Paper ↗
Get In Touch

Let's
Talk.

Reach out — I'd love to connect.