Knowledge of VMs and sandboxed execution environments (e.g., Xbox, Capsules).
About the role
Minimum qualifications:
Experience designing, building, and operating large-scale distributed systems, with a focus on scalability, fault tolerance, and reliability.
Experience in Machine Learning or AI infrastructure.
Experience in software performance analysis and improvement, and High-Performance computer architectures (e.g., CPU, memory subsystem, multiprocessing).
Experience designing software architecture for applications, including AI/ML systems.
Experience with site reliability engineering, system archiecture, and distributed systems.
Experience building developer tools to improve development velocity, code quality, or code health (e.g., CI/CD, compilers, test automation frameworks).
Preferred qualifications:
Master’s degree or PhD in Engineering, Computer Science, or a related technical field.
8 years of experience with data structures/algorithms, with strong proficiency in C++ or Python.
Experience developing and optimizing ML or production serving infrastructure.
Experience leading complex architectural migrations or cross-team infrastructure projects. Excellent problem-solving skills and ability to collaborate widely across organizations.
Knowledge of VMs and sandboxed execution environments (e.g., Xbox, Capsules).
Understanding of the latest evaluation research and areas important to LLM research.
Responsibilities
Lead the architectural design of high-throughput serving infrastructure for evaluation workflows.
Design scalable systems to store, process, and serve massive datasets for model evaluation.
Optimize the latency and resource efficiency of large-scale inference and evaluation pipelines.
Collaborate with research teams to define infrastructure requirements for next-generation AI models.
Establish technical roadmaps for data serving platforms supporting DeepMind's evaluation ecosystem.