Ziyi Zhang

Curriculum Vitae

Download PDF

Education

  • expected Jan 2027
    Ph.D. in Electrical and Computer Engineering, Carnegie Mellon University
  • 2022
    M.S. in Computer Science, Duke University
  • 2021
    B.S. in Mathematics & Computer Science, University of Michigan

Experience

  • Summer 2026
    Software Engineer Intern (Gmail Ads), Google LLC, Sunnyvale, CA
    • Built an LLM-based agentic system to adjust bidding parameters produced by the hidden distribution model for impressions.
    • By modeling bidding as a Markov Decision Process, the system aims to use the LLM’s reasoning ability to adjust the range of maximal bidding prices on a daily basis.
    • Used knowledge agent, critic agent, and Thompson sampling to select the most valuable knowledge on a weekly basis.
    • For the selected adgroups, the desired-spend-over-spend (DSoS) moved closer to 1 by about 0.03, with reduced variance in daily DSoS fluctuations.
  • Summer 2025
    AI System Intern, Aira Technologies Inc., Saratoga, CA
    • Designed and implemented an agentic LLM network for root cause analysis (RCA) in AT&T network.
    • Implemented an agentic system using LangGraph with a team of researcher LLMs each focusing on different elements of KPI, alarms, and tickets. The team is led by a supervisor LLM, which produces the final report.
    • Used GraphDB to build the knowledge graphs for the researcher LLMs to identify the root cause more accurately.
    • Improved the accuracy of RCA from 30% in the prototype version to 85% in the production version.
  • Summer 2020
    Software Engineer, Guochuang Software Co.
    • Designed a commercial website using HTML and JavaScript.
    • Redesigned the database structure in order to scale up the system’s capacity to handle larger number of transactions.

Publications

Conference Papers

  • Z. Zhang, S. Li, Y. Nakahira, P. Bogdan, and G. Qu. "HilbertGen-3D: Hilbert–Multifractal Conditioning for Topology-Aware 3D Generation." Advances in Neural Information Processing Systems (NeurIPS), 2026 (to appear). More
  • X. Zhang, Z. Zhang, Z. Zhao, S. Tu, G. Qu, Y. Nakahira, and P. Bogdan. "Learning Fractional-Order Dynamics from a Single Trajectory." Advances in Neural Information Processing Systems (NeurIPS), 2026 (to appear). More
  • X. Xu, Z. Zhang, Y. Nakahira, G. Qu, and Y. Chi. "Polynomial Convergence of Riemannian Diffusion Models." International Conference on Learning Representations (ICLR), 2026. More
  • Z. Zhang, Y. Nakahira, and G. Qu. "Stabilizing LTI Systems under Partial Observability: Sample Complexity and Fundamental Limits." Advances in Neural Information Processing Systems (NeurIPS), 2025. More
  • Z. Zhang, Y. Nakahira, and G. Qu. "Learning to Stabilize Unknown LTI Systems on a Single Trajectory under Stochastic Noise." In Proceedings of the Conference on Uncertainty in Artificial Intelligence (UAI), PMLR 286:4889–4919, 2025. More
  • Z. Zhang, G. Qu, and Y. Nakahira. "Fast Bandit-based Policy Adaptation in Diverse Environments." In Proceedings of the American Control Conference (ACC), pp. 2458–2464, 2025. More

Journal Articles

  • Z. Zhang, Y. Nakahira, and G. Qu. "Predictive Control and Regret Analysis of Non-Stationary MDP with Look-ahead Information." Transactions on Machine Learning Research (TMLR), 2025. More
  • Z. Zhang. "Polyhedra of small relative mixed volume." Beiträge zur Algebra und Geometrie / Contributions to Algebra and Geometry, 62(4):773–782, 2021. More
  • C. Bibby, A. Odesky, M. Wang, S. Wang, Z. Zhang, and H. Zheng. "Minimal flag triangulations of lower-dimensional manifolds." Involve, a Journal of Mathematics, 13(4):683–703, 2020. More

Preprints

  • Z. Zhang, X. Deng, G. Qu, and Y. Nakahira. "UAVGENT: A Language-Guided Distributed Control Framework." arXiv preprint arXiv:2602.13212, 2026. More

Teaching

  • 2024
    18-667: Algorithms for Large-scale Distributed Machine Learning and Optimization, Teaching Assistant
  • 2023
    10-625: Introduction to Convex Optimization, Teaching Assistant

Skills

Programming languages
C++ (3 years), Java (3 years), Python (7 years)
Databases
Oracle (1 year), SQL (1 year), MongoDB (1 year)
Math coursework
Stochastic Calculus, Real Analysis, Functional Analysis, Partial Differential Equations, Graph Theory
Data science
Statistical Analysis, Regression
Computer science
Numerical Methods, Reinforcement Learning, Deep Learning