Aug 2020 – Aug 2024 · 4 yrs · Gurugram, Haryana, India · Hybrid

• Data Scientist 1
• 1) Core developer of AI-powered Chargeback risk assessment score, part of a suite of 9 scores that bring $20M dollar revenue. Granted a product patent for the same work
• 2) Part of the 4 member team that came in top 10 in KDD Graph Data Challenge.
• 3) Built a Merchant Common Point of Compromise product, by formulating the detection of compromised merchant as a Graph learning problem.
• Data Scientist 2
• 1) Built the drift detection pipeline for maintaining model performance stability of Chargeback risk assessment score.
• 2) Focused on shifting from traditional XGBOOST models (trained using H2O) to Graph based models (dgl) for Chargeback Score, deployed for an entire region using TigerGraph Database and Kubernetes
• 2) Published an ECML-PKDD paper on Graphs for unsupervised representation learning, relevant for the same AI Scores suite of products
• 3) Developed end-to-end internal NL2SQL GenAI product by fine tuning CodeLLAMA and semantic search techniques.
• 4) Building Rules from a combination of the AI Scores and another suite of transaction level scores for capturing frauds at a transaction level.
• Senior Data Scientist
• 1) Focusing on building an internal AutoML framework to replace reliance on H2O (costs approx. $3M per annum)
• 2) Granted patents in the domain of Graphs, Chargeback, and AutoML
• 3) Currently focused on integrating GenAI tools especially NL2SQL (Code Llama) into AutoML systems for automated feature engineering/discovery

May 2018 – Jul 2018 · 2 mos · Pune, Maharashtra

• >Built a tool for the dynamic development of web apps on Angular 5
• >Integrated multiple components on UI with drag & drop functionality using NgX-DnD directory for a user-friendly experience in Web Development

Feb 2018 – May 2018 · 3 mos · India

• >One of 7 (of all 2nd and 3rd year) students from IIT Bombay to be selected for a semester-long program to learn Quantitative Trading Strategy Modelling by developing models in Python.
• >The program combined virtual training and assignments in developing alphas/trading models.
• >Worked on a Capstone Project and used the LSTM model to predict the movement of cointegrated stock prices.

May 2017 – Jul 2017 · 2 mos · Hyderabad Area, India

• > Analized and implemented five trading strategies on Indian equity market.
• > Completed a project on Mutual fund and proved that Indian mutual fund have been out performing the bench mark while most developed nation are unable to do so.
• > Explored the basics of finance by pursuing courses on Trading Algorithms from COURSERA, Bloomberg Market concepts to gain domain knowledge about the stock market and investment.

Jan 2017 – Jan 2020 · 3 yrs · Mumbai Metropolitan Region

• >Appointed as Teaching Assistant for Complex Analysis under Prof. Sudarshan R. Gurjar, Dept. of Mathematics, IIT Bombay, Introduction to Machine Learning and Advanced Machine Learning under Prof. Amit Sethi, Dept. of Electrical Engineering, IIT Bombay.
• >Assisted 50+ juniors by taking weekly tutorial sessions and additional help sessions