Sep 2023 – Aug 2024 · 11 mos · Hyderabad, Telangana, India · On-site

Dec 2022 – Dec 2023 · 1 yr · Gurugram, Haryana, India

• Fraud Management and Customer De-Duplication
• ∗ Developed clustering of similar addresses on a dataset of over 550+ million records, measured by a remarkable response time of 50ms and 1000 TPS using Golang and the BadgerDB KV store.
• ∗ Developed rule-based customer deduplication REST API in Golang, handling 5 million daily hits, achieving 500 TPS throughput and 200 ms response time by integrating Postgres and dedicated gRPC service for address matching.
• ∗ Optimized performance through algorithm optimization and utilization of advanced data structures like roaring bitmap, bloom filters, priority queue, and Misra–Gries counting algorithm from 2s 100 TPS to 50ms 1000 TPS.
• ∗ Developed UI for business users to define matching rules to enable real-time rule customisation, eliminate the API downtime and reduce rule change duration from weeks to real-time in React.
• ∗ Achieved a notable improvement in targeted marketing by identifying customers in one LOB, resulting in a 50% increase in conversion rates by effectively targeting customers interested in multiple services across different LOBs.
• ∗ Developed fraud management, identifying the fraud of avg 5 million Rupees per month using customer deduplication API in Golang.
• ◦ Inventory Management System :
• ∗ Designed and developed a centralized platform for tracking inventory across multiple locations and systems, streamlining inventory management, reducing errors, improving efficiency, and saving costs.
• ∗ Implemented webassembly code in C++ for image preprocessing and barcode scan optimization, resulting in 10x faster scanning.
• ∗ Technologies Used: Golang, Hasura, PostgreSQL, Graphql, Docker, Kubernetes, and ReactJS

Jun 2021 – Dec 2022 · 1 yr 6 mos · Gurugram, Haryana, India

Jan 2020 – May 2021 · 1 yr 4 mos · Hybrid

• Worked on developing a legal text mining system for the common law house (LTH) funded by Calcutta High Court.