Raghav Apoorv

Built CI/CD and parity workflows for metric generation—99.9% deployment reliability. Led the competitive metric modeling library that cut dev time in half and runtime by two-thirds. Built ML models predicting rider-to-eats conversion and driver behavior. Designed a Spark/Hive data platform for driver analytics and migrated legacy Presto pipelines to Spark.

Led plagiarism and cheating detection for CS6601 (Artificial Intelligence) for nine terms. With Prof. Thad Starner and the Jack Watson Research Group, developed Examinator—a tool that detects cheating in take-home and online exams, reviewing 2M+ submission pairs and resulting in 56+ referrals. Supported 2000+ students via Ed and office hours. Received a Thank a Teacher Award.

Joined as a software engineer and was quickly promoted to Tech Lead Manager. Owned the design and delivery of an in-house chat platform (Golang, SQS, Redis, MySQL) that replaced a third-party tool and scaled to 60× the previous load—managed a team building React Native SDKs for Android and iOS, a 500+ agent management system, and a decision-tree chatbot that cut support queries by 60%. Led the monolith-to-microservices migration in Go, driving architecture decisions and coordinating across teams. Owned digital ads targeting, leveraging data science models to save ~$500K monthly, and led growth initiatives that increased the user base 10% month over month.

Built a hyperlocal super-aggregator that combined cab services, restaurants, flights, and e-commerce price comparison into one app. The core technology was acquired and used by Micromax in their "Around YU" product.

Improved the regulatory stress testing system, reducing runtime by ~30%. Built a tool in Java to migrate projects from Ant to Maven and worked on migrating the code repository from SVN to Git.