Interested in robotics, artificial intelligence, machine learning, and broad applications of said topics to a variety of fields (e.g. medicine, infrastructure, etc.)!
Experienced tutor with a demonstrated history of working in the higher education industry. Skilled in Computer Science and Mathematics.
Worked as an SDE I in the GSF Capacity Management and Flex Offer Marketplace and Optimization teams.
* Amazon Restaurants India Instant Offers Launch: Coordinated multiple development teams (mobile, pricing, routing), business ops, and Indian site managers over the course of 3 months to launch a new gig delivery method for Amazon Restaurants across all India sites. This reduced Cost Per Delivery by 86% from 317 rupees to 45.3 rupees without degradation of customer delivery time or performance.
* Low-Lead Offer Targeting and Assignment: Designed and developed an experimental low-lead offer targeting system using AWS to periodically assign personalized offers to Amazon Flex delivery drivers. This aimed to increase driver engagement and reduce immediate labor shortages whenever packages to be delivered in the next hour lacked drivers.
* Instant Offers Acceptance Rate Machine Learning Improvements: Improved precision for a Machine Learning model for predicting driver acceptance probabilities for personalized offers from 61% to 72%. Created an automated model training pipeline by reducing training data processing memory and runtime through Pandas.
* Late Order Notification Vending System: Assisted in designing and implementing an ETA and confidence vending system, including responsibilities like sending out notifications for late order deliveries.
* Transportation Capacity Redshift Database Deprecation: Coordinated with developer and business analytics teams to audit and deprecate unused portions of a 3-year old AWS Redshift database inappropriately used by other teams. Cluster size and annual cost were reduced by 50% and annual ticket volume by 100%.
* Sprint Process Improvements: Utilized new metrics and tools to our team's Sprint Process to increase daily stand-up accountability of long-standing issues and facilitate communication in addressing blockers while simultaneously reducing overall weekly meeting times.
Worked as an SDE Intern for the Last Mile Flex Scheduling team.
* Delivery Scheduling Data Consistency: Analyzed production data and designed new data schema using AWS S3, Lambda, and ElastiCache to resolve data consistency issues in response to database updates.
* Robotics Control: Worked under Oussama Khatib to develop a Java semantic parsing server integrated with a C++ operational space robot controller for flexible, natural control of a variety of robot arms.
* Natural Language Processing: Implemented a Python Redis Publish/Subscribe script to communicate and process information from a modified Natural Language Processing Javascript web client.
