Rohit Shukla

As a Data Scientist at Mastercard, I reduced the runtime of the merchant deduplication model by 85% and saved $800K annually by deploying a GPT model for transaction data cleansing. I also enhanced fraud detection accuracy by 16% through a unique merchant ID creation model.

At Recordsure, I automated topic annotation by training and fine-tuning BART, which reduced the project timeline by 30%.

In my role as a Junior Data Scientist, I deployed a string similarity recommendation model achieving ~94% accuracy and optimized it to reduce runtime significantly. I also trained and deployed an NER model for contract entity extraction.

As a Procurement Analyst, I implemented an ARIMA model to predict copper prices with 92% accuracy, saving the company approximately $2.6M annually.

In my position as a New Product Developer, I extracted and analyzed supplier and inventory data using MySQL and Power BI, optimizing the supply chain for 16 products and saving approximately $6M annually.

Completed a Master's degree in Data Science and Analytics with a thesis focused on large scale graph community detection algorithms.

Earned a Bachelor's degree in Mechanical Engineering, focusing on engineering principles and applications.