# Fangyi Zhu > Research Software Engineer Location: San Francisco Bay Area, United States Profile: https://flows.cv/fangyi I'm open to Research Software Engineering roles! Conducting research "Finding Interpretable Features of Depression Symptoms in Large Language Models" at Stanford Precision Neurotherapeutics Lab. My PIs are Ajay Subramanian, PhD and Professor Corey Keller, MD, PhD. Our tech stack is PyTorch, SciPy, Pandas, Numpy, scikit-bio, and Plotly. Check my Github for updates! ## Work Experience ### Software Engineer @ Google Jan 2022 – Jan 2024 | San Francisco Bay Area ❏ Designed and developed a software quality monitoring and bug tracing tool with Google SQL, Python, HTML, JavaScript, Spanner Database, Git-on-Borg, GCP Big Table and Data Visualization. The tool enables hundreds of Android and Pixel developers to trace quality and test performance issues back to specific code reviews, authors, and lines of code. ❏ The tool, which is used for internal and external Git-on-Borg repositories, consolidates data from seven databases, including code changes, code reviews, code line coverage, ticketing systems, test runs, test logs, and test result data. The tool then presents this data in an interactive data visualization format, allowing users to search and trace quality issues, performance issues, and bugs through drilldown functionalities. ❏ Guided a summer intern in developing the essential features of the code quality tool, ensuring the internship's successful and punctual completion. ❏ Established consistent data pipelines and an interactive dashboard utilizing SQL, Python, Google SQL and Google Cloud BigTable. The pipeline and dashboards facilitated the analysis of test results data by Nest developer and testing teams, encompassing hundreds of test builds and test runs across Nest on a daily basis. ❏ Attended internal and external computing conferences including Grace Hopper Celebration and spoke at knowledge sharing sessions at Google, hosted team events. ### Software Development Engineer @ Amazon Jan 2017 – Jan 2021 | Seattle, Washington, United States ❏ Designed, developed and deployed GDPR-compliant advertisement products for IMDb across various platforms like desktop, mobile, Android, and iOS, utilizing AWS Dynamo DB, the Amazon Advertising Platform, Java, Swift, Android Studio, and ReactJS. ❏ These products provide detailed user interaction tracking, allowing internal and external advertisers including Amazon Retail and major movie studios to monitor impressions, click-through rates, and conversion rates. The advertising products are prominently displayed on IMDb's movie and celebrity pages, as well as the front page, reaching millions of users daily. ❏ Built internal tools and advertisement templates utilizing React JS, Java, HTML and AWS. These tools enable IMDb advertisement designers and developers to streamline the advertisement design, troubleshooting, and deployment processes. The major movie studios can thus quickly iterate through the design process with IMDb, receiving visually appealing and custom-tailored advertisements for their movies. ❏ Designed a machine learning model using Amazon’s Personalization platform to find similar movie titles and mentored a summer intern to implement it. ❏ Identified 2.5 times more IMDb users from Amazon traffic by integrating a Java lookup component into the Amazon Retail website. IMDb can now detect and predict the movie preferences of millions of additional Amazon users by analyzing large amounts of user data from Amazon using AWS data services like S3, EMR, and Dynamo DB. ❏ Led a team to win a company-wide hackathon by developing software using Python and AWS Rekognition’s facial recognition to identify and track celebrities in movies and videos. ### Software Engineer @ Point Inside Jan 2016 – Jan 2016 | Bellevue ❏ Developed a reliable continuous deployment system that streamlines the testing and code deployment processes for AWS applications. This solution empowers application teams to manage highly scalable web applications utilizing Jenkins and Python. Millions of shoppers have navigated through stores like Lowe’s, The Home Depot, Staples, and Costco with the help of our indoor maps. These maps guide customers to the precise shelves where their desired products are located. ❏ Designed and implemented AWS Lambda functions to automate the management of AWS resources, optimizing monitoring and achieving significant cost savings. These functions resulted in a reduction of AWS costs by 15%. ❏ Wrote Chef cookbooks that deploys fully automated, self-healing MongoDB replica sets with AWS Autoscaling Group. These MongoDB replica sets guaranteed consistent database availability, which was crucial during the holiday shopping season, when shopping traffic surge. (Open-sourced on GitHub) ❏ Mentored engineers about cost-saving strategies for AWS EMR and Data Pipeline. ### Software Development Engineer Intern @ Amazon Web Services Jan 2014 – Jan 2014 ❏ Using Amazon EMR (Elastic MapReduce) and Data Pipeline, built a Hadoop application on Amazon Elastic File System which analyzed terabytes of Wikipedia data. This application served as a daily stress test for Amazon Elastic File System prior to its official release. ### Software Development Engineer Intern @ Amazon Jan 2013 – Jan 2013 ❏ Implemented and tested core Java APIs and database management function for Amazon financing option service. Starting from Brazil, Amazon financing option service now serves millions of Amazon customers around the world. ## Education ### Bachelor of Science (B.S.) in Computer Science University of Minnesota Jan 2011 – Jan 2014 ### Foundational Modules Graduate Amazon Machine Learning University Jan 2018 – Jan 2018 ### Quantum Computing Fundamentals MIT xPRO Jan 2021 – Jan 2021 ## Contact & Social - LinkedIn: https://linkedin.com/in/fangyizhu - GitHub: https://github.com/fangyizhu --- Source: https://flows.cv/fangyi JSON Resume: https://flows.cv/fangyi/resume.json Last updated: 2026-03-22