Hailong Xiao

About Hailong Xiao

I am a distributed systems and ML infrastructure engineer with 9+ years of experience building and scaling large-scale production systems at Google.My work focuses on enabling machine learning workloads to run reliably and efficiently at fleet scale. I have led platform adaptations for next-generation hardware, designed resource management frameworks for cluster scheduling, and delivered systemic reliability improvements that unblock production ML jobs.In recent years, I have become increasingly interested in the intersection of large-scale systems and the mathematical structure of modern LLMs. I am particularly drawn to problems involving attention efficiency, KV cache optimization, distributed training stability, and resource-aware model architectures.With a background in computational and applied mathematics (PhD), I am motivated by problems where mathematical insight and systems design meet — improving performance, stability, and scalability of large models in production environments.Areas of expertise include: Large-scale distributed systems and cluster scheduling ML infrastructure and production reliability Resource management and fleet-scale observability Ads review and integrity systems

Experience

1. Software Engineer

   Google

   Mar 2016 — Present · US

   Cloud Platforms Infrastructure Engineering (PIE)(2023 – Present)• Leading Google’s fleet-wide ML infrastructure and cluster management strategy, driving reliability, scalability, and next-generation hardware adoption.• Defined multi-year ML infrastructure strategy, enabling Arm-based TPU hosts and future-proofing Google’s specialized hardware stack.• Improved global fleet reliability and resource utilization through systemic failure mitigation frameworks and automated orchestration- Google Assistant (Bard / Gemini)(2021 – 2023)• Led Gemini App Extensions platform, enabling integrations with Maps, Workspace, Hotels, and Flights.• Scaled Smatspace weather pipelines to 100% user coverage, improving availability and operational observability.• Drove cross-team engineering improvements and observability frameworks to support large-scale system migrations- Ads Integrity (2016 – 2021)• Developed scalable, reliable systems to enforce ad policy and maintain platform integrity.• Built high-throughput pipelines for policy enforcement and ad content retrieval.• Ensured platform safety and compliance at global scale.

Education

• The University of Texas at Austin

  Master of Science (M.S.), Computational Science, Engineering and Mathematics

  2008 — 2011

• The University of Texas at Austin

  Doctor of Philosophy (PhD), Computational and Applied Mathematics

  2008 — 2013

• University of Science and Technology of China

  Bachelor of Science (BS), Pure and Applied Mathematics

  2004 — 2008

Skills

• Mathematical Modeling
• Finite Element Analysis
• Linux
• Matlab
• C
• Seismic Imaging
• Finite Difference Method
• Parallel Computing
• Mpi
• Computational Physics
• Simulations
• Modeling
• Cfd
• Latex
• Mathematica
• Statistics
• C++
• Probability
• Numerical Simulation
• Scientific Computing
• Programming
• Algorithms
• Python