Portfolio

About Me

I’m a Senior Data Scientist based in the Bay Area, building production ML systems at the intersection of risk intelligence and applied AI. I’m hacker/developer at heart, love to get my hands dirty at interesting problems.

At BILL, I work on credit risk models, agentic LLM workflows, and semantic search systems that drive real business outcomes — from $5M+ in revenue uplift to automating complex underwriting processes end-to-end.

Before BILL, I spent 3 years at American Express working across credit modeling and risk strategy for Global Collections and US Commercial teams — where I learned to build models that matter at scale.

I’m currently exploring new opportunities in Data Science (Product, Finance, Strategic Intelligence), MLE, and Applied AI roles in the Bay Area. Open to connecting and collaborating.

Download Resume

Latest Work

CineSphere - Movie Recommending Chatbot Using Knowledge Graphs

Developed a chatbot using GPT4 with multi-agent LangChain Framework and Knowledge Graph over 45K movies
Leveraged OpenAI Function Calling, Vector Indexing, and Node similarity algorithms to create robust chat interface and recommendations

Developed a classic Encoder-Decoder architecture using ConvLSTM and LSTM for weather forecasting at Lake Michigan
Used satellite images and meteorological data to create the model attaining a recall of 0.85 on a highly imbalanced data

TravelBud - AI Tool for Travel planning

Developed a travel itinerary-generating application optimized based on user areas of interest, budget, and destination
Leveraged RapidAPI for flight, hotel data, and OPENAI GPT3 for chat completion coupled in a cloud ecosystem

A/B Test — Understanding causal relationships in experiments

Published an article on A/B tests to understand implementation and how they can help understand causal relationships in experiments.

DIME(Database on Ideology, Money in Politics, and Elections) Network Analysis

Researched DIME data to formulate a network among the monetary contributions in US Federal Elections
Leveraged network science principles to showcase scale-free properties, and ideology-based communities among contributors

Founder - GrocerEase: Smart Digitization of Groceries

GrocerEase is a new-age startup helping its users digitize the process of tracking groceries efficiently and simply. We address a very common problem of forgetfulness among grocery shoppers, which results in redundant purchases and food wastage quite often.
The user just needs to click a picture of the grocery purchase receipt, and the app automatically digitizes the items, along with their quantity in a tabular format, and additionally estimates a tentative expiry date based on the internal database to give out expiry alerts, such that the product can be used before it’s about to expire.We use Deep learning and Computer Vision under the hood to develop our product
Finished among the Top 5 finalists competing for $10K seed funding in the Husky Startup challenge at Northeastern University (A student-led startup incubator by the Entrepreneurship Cell)

VaniVerse

VaniVerse is a conversational AI platform that can be used to generate prompts for a given audio file
OPENAI Whisper & GPT API are leveraged to process audio files and generate prompts simultaneously using Airflow

Hindi Seq2Seq Model - Generating Hindi poems with RNN

Trained a Sequence Model on a corpus of Hindi Poems to generate new sentences leveraging NLP, text processing principles
Used pre-trained tokenizing and word embeddings for the Hindi language developed by AI4Bharat

Runner’s High: Reflecting on Runs in ‘22

In 2022, I took on running as a hobby and set a target of running 500km over the calendar year. Reflecting on the runs, I used Python to visualize the geospatial data collected over the year on Strava, sharing some thoughts and insights!

Feel Free to email -