{"schemaVersion":"jobsearcher.job.v1","id":"eeaf76f2a223ba7bcc5b05e9","url":"https://jobsearcher.com/jobs/eeaf76f2a223ba7bcc5b05e9","canonicalUrl":"https://jobsearcher.com/jobs/eeaf76f2a223ba7bcc5b05e9","title":"Software Engineer, Hardware Accelerators Performance, GeminiApp, DeepMind","description":"Overview\nAs a Software Engineer at Google DeepMind, you will advance Gemini training and serving efficiency on TPUs and GPUs. You will profile and optimize large-scale distributed workloads and design high-performance kernels to maximize hardware utilization. You will build automated benchmarks and diagnostics to guard against regressions and help shape accelerator architectures and compiler roadmaps. You join a mission-driven team working at the intersection of modeling, compilers, and infrastructure to deliver end-to-end efficiency.\n\nCompensation / Benefitsbonus targetequitybenefitscompetitive compensationcareer development opportunitieshealth and well-being programs\nResponsibilitiesImprove training and serving efficiency of Gemini models on TPUs and GPUs across configurations and runtimesProfile large-scale distributed workloads to diagnose compute, memory, and communication bottlenecks and identify optimization opportunitiesDesign and implement high-performance, low-level custom kernels for critical model operatorsBuild and maintain automated benchmarking pipelines and diagnostic tooling to track, reproduce, and guard against performance regressionsInfluence accelerator architectures and compiler roadmaps by feeding back workload profiles and model requirements\nKey requirementsBachelor’s degree or equivalent practical experience2 years of experience with C++ or PythonExperience with profiling tools (gProf, Valgrind, or VTune) for performance optimizationExperience in software development for hardware accelerators (CPUs, GPUs, or TPUs), including memory-hierarchy or instruction-level tuningstrong software engineering foundationenjoy solving challenging problemscollaborative and mission-drivenprofiling tools (gProf, Valgrind, VTune)performance optimizationsoftware development for hardware accelerators (CPUs, GPUs, TPUs) including memory-hierarchy or instruction-level tuning","company":"Google","rawCompany":"google","city":"San Jose","state":"CA","isRemote":false,"isActive":false,"createdAt":"2026-09-15T03:28:37.685Z","occupations":[{"code":"15-1252.00","title":"Software Developers","slug":"software-developers"},{"code":"15-1299.08","title":"Computer Systems Engineers/Architects","slug":"computer-systems-engineers-architects"},{"code":"17-2061.00","title":"Computer Hardware Engineers","slug":"computer-hardware-engineers"}],"industries":[{"code":"513210","title":"Software Publishers","slug":"software-publishers"},{"code":"541511","title":"Custom Computer Programming Services","slug":"custom-computer-programming-services"},{"code":"541512","title":"Computer Systems Design Services","slug":"computer-systems-design-services"}],"jobPosting":{"@context":"https://schema.org","@type":"JobPosting","title":"Software Engineer, Hardware Accelerators Performance, GeminiApp, DeepMind","description":"Overview\nAs a Software Engineer at Google DeepMind, you will advance Gemini training and serving efficiency on TPUs and GPUs. You will profile and optimize large-scale distributed workloads and design high-performance kernels to maximize hardware utilization. You will build automated benchmarks and diagnostics to guard against regressions and help shape accelerator architectures and compiler roadmaps. You join a mission-driven team working at the intersection of modeling, compilers, and infrastructure to deliver end-to-end efficiency.\n\nCompensation / Benefitsbonus targetequitybenefitscompetitive compensationcareer development opportunitieshealth and well-being programs\nResponsibilitiesImprove training and serving efficiency of Gemini models on TPUs and GPUs across configurations and runtimesProfile large-scale distributed workloads to diagnose compute, memory, and communication bottlenecks and identify optimization opportunitiesDesign and implement high-performance, low-level custom kernels for critical model operatorsBuild and maintain automated benchmarking pipelines and diagnostic tooling to track, reproduce, and guard against performance regressionsInfluence accelerator architectures and compiler roadmaps by feeding back workload profiles and model requirements\nKey requirementsBachelor’s degree or equivalent practical experience2 years of experience with C++ or PythonExperience with profiling tools (gProf, Valgrind, or VTune) for performance optimizationExperience in software development for hardware accelerators (CPUs, GPUs, or TPUs), including memory-hierarchy or instruction-level tuningstrong software engineering foundationenjoy solving challenging problemscollaborative and mission-drivenprofiling tools (gProf, Valgrind, VTune)performance optimizationsoftware development for hardware accelerators (CPUs, GPUs, TPUs) including memory-hierarchy or instruction-level tuning","datePosted":"2026-09-15T03:28:37.685Z","dateModified":"2026-09-15T03:28:37.685Z","hiringOrganization":{"@type":"Organization","name":"Google","sameAs":"https://jobsearcher.com"},"jobLocation":{"@type":"Place","address":{"@type":"PostalAddress","addressLocality":"San Jose","addressRegion":"CA","addressCountry":"US"}},"identifier":{"@type":"PropertyValue","name":"JobSearcher","value":"eeaf76f2a223ba7bcc5b05e9"},"url":"https://jobsearcher.com/jobs/eeaf76f2a223ba7bcc5b05e9"}}