Tony Tu

Worked on applied reasoning and the long-horizon credit assignment problem Designed a divergence-based hierarchical reinforcement learning framework to attribute reward signals across extended decision sequences

AI Applied Scientist Intern at Zillow (2026-05 – 2026-08)

• Worked on applied reasoning and the long-horizon credit assignment problem
• Designed a divergence-based hierarchical reinforcement learning framework to attribute reward signals across extended decision sequences

Software Engineering Intern at Meta Inc. (2022-05 – 2022-08)

• Worked with Ads fairness equity team to design and train reinforcement learning models to address potential bias in current Meta ads delivery system
• Used policy gradient network to implement fine-grained RL controller to further improve ad delivery to users with different backgrounds including gender, age and ethnicity

Head GTA at Georgia Institute of Technology ECE Department (2023-08 – Present)

• Head graduate teaching assistant for ECE 4180 - Embedded System Design
• Hold regular 10 weekly lab hours and help students with completing assignments and understanding of lab materials

Head uGSI at UC Berkeley EECS Department (2018-04 – 2019-04)

• Head uGSI for Engineering 98 for 3 semesters, created course contents, lecture recordings and autograders for 60+ students
• Made and maintained a fully functional course website for the Summer 2018 and Winter 2019 Semester: (hosted on Heroku):

Ph.D. in Machine Learning – Georgia Institute of Technology (2023-08)

M.S. in Computer Science – Georgia Institute of Technology (2021-08 – 2023-05)

B.S. in Electrical Engineering and Computer Science – University of California, Berkeley (2017-08 – 2021-05)