May 2022 – Sep 2024 · 2 yrs 4 mos · Bengaluru, Karnataka, India · Hybrid

• Developed two boosting models to detect fraud and identify high-value customers during onboarding, achieving automation in decision-making and improving business metrics such as time to open accounts (TTO) and deposits.
• Developed an XGBoost model to identify key customers essential for growth, enhancing Google’s targeting efforts. This solution increased core customer numbers and speed up account openings.
• Led a team through development, validation, and deployment of a credit solutions suite, creating models to boost engagement and reduce default rates.

Nov 2021 – May 2022 · 6 mos · Remote

• Share of Wallet:
• Developed a heuristic model to estimate merchant share of wallet at demographic level, providing market penetration insights through a fine-tuned POC model.

Apr 2020 – Nov 2021 · 1 yr 7 mos · Remote

• Merchant Matching Solution: Optimized an existing algorithm to link records by name and demographics, increasing record coverage and match accuracy while improving performance and reducing computation time.
• Granular Cluster Optimization: Validated and implemented a graph-based clustering approach with NetworkX to reduce over-linking errors, enhancing clustering precision with minor trade-offs in under-linking.

Jul 2018 – Apr 2020 · 1 yr 9 mos · Mumbai Metropolitan Region · On-site

• ChatBot: Developed a Virtual Agent for a university using DialogFlow, significantly increasing user engagement. Enabled instant responses to queries and seamless live agent scheduling.
• Human Disease Prediction: Built a POC to predict diseases based on symptoms and context, supporting early treatment and cost reduction. Used transformer model embeddings for effective symptom interpretation.
• Engine Defect Detection: Designed a POC using Deep Learning to identify engine defects in car cylinders, applying YOLO for object localization and SqueezeNet for defect classification. Employed OpenCV data augmentation to handle class imbalance and enhance model accuracy.
• Leadership Skills: Trained and assessed new Machine Learning Engineers to ensure they met company standards .