Jun 2022 – Jul 2025 · 3 yrs 1 mo · Bengaluru, Karnataka, India · Remote

• Increased monthly recoveries by 8.9% by leveraging A/Btesting.
• Reduced net credit loss by 1.2% by providing long-term relief strategies to applicable merchants.
• Created customer segmentation model by leveraging statistical analysis and machine learning models (DecisionTree).
• Worked on end-to-end model building in credit domain to predict the expected repayment from a customer against their loan.
• Built an offline machine learning model to predict customer’s probability of becoming default in a specific time frame.

Feb 2022 – Jun 2022 · 4 mos · Remote · Remote

• Improved the current ROAS prediction pipeline by a factor of 50% by analyzing marketing data for clients as well as competitors.
• Leveraged state-of-the-art models like Robyn (by Meta) to solve channel attribution problem.
• Utilised Shapley, Adstock, and market saturation to work on Marketing Mix Modeling (MMM).

Aug 2021 – Jan 2022 · 5 mos · Noida, Uttar Pradesh, India · Remote

• Reduced the word matching algorithm runtime by 96% by building a custom
• Hashmap-based indexer to store similar words in the same key.
• Increased the OCR accuracy by 25% by incorporating an image straightening algorithm before feeding the images in OCR.
• Improved word suggestion algorithm’s performance by 20% by returning the most similar word based on NLP and edit-distance among all the suggested words.

Jan 2020 – Aug 2021 · 1 yr 7 mos · Gurugram, Haryana, India · On-site

• Reduced machine learning (data) pipeline implementation time by 25% by designing efficient and robust ETL (Extract Transform Load) workflows with 1000+ lines of code and 20+ modules.
• The phases in the data pipeline consisted of importing, cleaning, transforming, and manipulating data, as well as training machine learning models and hyper-parameter optimization.
• Increased client's sales by 20% by generating cost-curves to identify saturation points for each advertising channel for their marketing campaigns.