I am currently a Software Engineer at DoorDash on the Logistics / ETA team, where I build modelling systems that estimate end-to-end delivery times across the platform. Before joining DoorDash, I earned an M.S. in Machine Learning from Columbia University, where I developed a strong interest in machine learning and large-scale distributed systems. When the weather's nice, you can usually find me playing pickleball, frisbee, or bouldering; otherwise, I'm likely reading a new book or playing poker with friends. I'm always excited to learn from well-designed systems, collaborate with great engineers, and tackle challenging technical problems.
Contact: alina.ying@gmail.com
I am a Software Engineer at DoorDash on the Logistics / ETA team, where we develop machine learning systems that estimate end-to-end delivery times across the platform. My work spans improving pre-checkout ETA accuracy through feature engineering and model iteration, as well as developing an explainable post-checkout pipeline that continuously updates ETAs as deliveries progress.
I was a Software Engineer at Microsoft on the Azure App Service team, where I built cloud computing infrastructure that powers serverless platforms such as Azure Functions and Azure Logic Apps. My work focused on autoscaling virtual machines based on workload demand, load balancing traffic, and improving the reliability, availability, latency, and performance of the data plane.
I was a Software Engineer Intern on the Uber Transit team, where I helped build integrations with public transit agencies to bring multimodal transportation options into the Uber app, making it easier for riders to plan and complete trips that combine public transit with Uber services.
LinkedIn Github Devpost
I looked into the acoustic-prosodic and lexical correlates of charisma in different speaker and rater demographic and personality groups, and in different speech genres. The dataset used were speech stubs from the 2019-2020 candidates for Democratic nomination for U.S. president.
I worked on path-planning with snake robotics in the ROAM Lab under Professor Matei Ciocarlie, focusing on using rapidly-exploring random trees (RRTs) to frontload path-planning computation. Experiments took into account dry, viscous, and fluid friction and were simulated on 5-link, 10-link, and 15-link snakes.
I worked on income inequality simulation in the CRIS Lab under Professor Venkat Venkatasubramanian, modeling income inequality reflected in wage distribution for a population of different classes (eg 90%, 9%, 1% or 50%, 30%, 15%, 5%) based on utility functions of compensation (alpha), effort (beta) and competition (gamma). I visualized the relative distributions and their sensitivities to alpha, beta, and gamma values after optimization.
Columbia University, Master of Science (MS), Machine Learning, 2021-2023.
Columbia University, Bachelor of Science (BS), Computer Science, 2017-2021.
During my time at Columbia, I was otherwise involved in: