Yifan Mai

Software engineer with 12 years professional experience

Skills: Full-stack web development, cloud infrastructure, CI/CD, developer tools, pipeline orchestration, distributed data processing, open source software maintenance, AI/ML productionization, LLM evaluation

Languages: Python, TypeScript, JavaScript, C++, Scala

B.S. Computer Science 2014, Stanford University

M.S. Computer Science 2014, Stanford University

Research Engineer, Stanford Center for Research on Foundation Models (CRFM)

September 2022 – July 2026

• Led development of the open source crfm-helm Python library for LLM and multimodal model evaluation, which has 2.9k stars on GitHub and was used by Apple, Meta, Microsoft and IBM
• Built benchmarks and leaderboards for LLM and multimodal evaluation across aspects including safety, multi-linguality, and biomedical-domain tasks
• Co-authored AI research publications in leading venues including NeurIPS, ICLR, ICML, JMLR and ACL
• Collaborated with MLCommon on the development of the AILuminate benchmark for safety testing for general purpose chat systems

Senior Software Engineer, Google

January 2019 – September 2022

• Maintained TensorFlow Extended (TFX), a platform for production machine learning pipelines
• Improved efficiency, performance and scalability of Cloud Dataflow distributed data processing, reducing the cost of running data processing pipelines with many parallel GroupByKeys by 50%
• Designed and implemented components and APIs for ingesting data from internal data stores into TFX
• Promoted internal adoption of TFX by running introductory courses and providing technical support to users

Senior Software Engineer, Coursera

August 2011 – March 2012 & June 2014 – January 2019

• One of three founding software engineers
• Designed and implemented the first iteration of Coursera's learning platform on a three-month timeline, working specifically on the web video player, course authoring tools, learner analytics dashboards, and web accessibility
• Implemented the second iteration of Coursera's discussion forums, which included user roles, full-text search, sub-forum hierarchies, content moderation, and content flagging
• Reduced development environment setup time for new engineers from five days to half a day by migrating from VMs to cloud instances and automating image building using Packer
• Designed and implemented a request-specific dynamic routing system for engineers to QA canary clusters of backend services in the production environment using Envoy service mesh
• Developed disaster recovery procedures to restore all online production systems to AWS cloud infrastructure from backup within 48 hours in the event of catastrophic system failure