{"schemaVersion":"jobsearcher.job.v1","id":"5214ff3a92227df3211d3814","url":"https://jobsearcher.com/jobs/5214ff3a92227df3211d3814","canonicalUrl":"https://jobsearcher.com/jobs/5214ff3a92227df3211d3814","title":"KERNEL ENGINEER","description":"ABOUT THE COMPANY\r\nWe're building autonomous research agents for recursive self-improvement (multi-agent systems that propose, run, and analyze machine learning experiments). We're a small team based in San Francisco, on-site\r\nABOUT THE ROLE\r\nYou'll write and optimize the GPU kernels and supporting systems software that makes our training and inference workloads fast. This is deep, low-level work (performance counters, memory bandwidth, warp-level scheduling) applied to the specific shapes and patterns our models actually use.\r\nWe hire kernel engineers because the gap between \"this works\" and \"this is fast on the hardware we have\" is enormous, and that gap directly bounds what our researchers can try. You'll close that gap.\r\nWHAT YOU'LL DO\r\nWrite and optimize GPU kernels (CUDA, ROCm, Triton, or similar) for training and inference workloads: attention variants, MoE layers, custom activations, communication primitives\r\nProfile real workloads with hardware counters and translate findings into specific kernel-level optimizations\r\nCo-design kernels with the research teams, when the kernel and the algorithm need to change together, you participate in both\r\nIntegrate optimized kernels into our training and serving stacks; benchmark before and after; verify the win is real end-to-end\r\nMaintain kernel quality over time as hardware, frameworks, and workloads shift underneath\r\nSpread kernel-level fluency across the team; we want this expertise shared, not siloed\r\nWHAT WE'RE LOOKING FOR\r\n4+ years writing performant GPU kernels (CUDA, ROCm, Triton, or production-grade equivalent)\r\nHardware-level fluency: memory hierarchy, occupancy, register pressure, tensor cores, warp scheduling\r\nProfiling fluency (Nsight, ncu, or comparable tools) and the discipline to measure before changing\r\nTrack record of shipping kernel-level optimizations that moved a measurable metric in a real system\r\nStrong systems expertise: you understand how kernels live inside larger frameworks and how integration choices affect end-to-end performance\r\nComfortable reading framework-level Python and C++ around your kernels\r\nNICE TO HAVE\r\nOpen-source contributions to kernel libraries, compilers, or ML frameworks\r\nExperience with multiple accelerator architectures (different GPU families, TPUs, custom ASICs), preferably AMD GPUs\r\nFamiliarity with collective communication primitives (NCCL or equivalent)\r\nCompiler or runtime background\r\nTHIS ROLE IS PROBABLY NOT FOR YOU IF\r\nYou haven't gotten your hands dirty at the kernel level: this isn't a higher-level systems role rebranded\r\nYou want to stay narrowly in one library; we expect breadth across the kernel surface our models actually use\r\nPerformance work without measurable end-to-end impact frustrates you\r\nJ-18808-Ljbffr","company":"Makermakerai","rawCompany":"makermakerai","city":"Millbrae","state":"CA","isRemote":false,"isActive":false,"createdAt":"2026-06-25T00:42:32.885Z","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":"15-1221.00","title":"Computer and Information Research Scientists","slug":"computer-and-information-research-scientists"}],"industries":[{"code":"513210","title":"Software Publishers","slug":"software-publishers"},{"code":"541511","title":"Custom Computer Programming Services","slug":"custom-computer-programming-services"},{"code":"541715","title":"Research and Development in the Physical, Engineering, and Life Sciences (except Nanotechnology and Biotechnology)","slug":"research-and-development-in-the-physical-engineering-and-life-sciences-except-nanotechnology-and-biotechnology"}],"jobPosting":{"@context":"https://schema.org","@type":"JobPosting","title":"KERNEL ENGINEER","description":"ABOUT THE COMPANY\r\nWe're building autonomous research agents for recursive self-improvement (multi-agent systems that propose, run, and analyze machine learning experiments). We're a small team based in San Francisco, on-site\r\nABOUT THE ROLE\r\nYou'll write and optimize the GPU kernels and supporting systems software that makes our training and inference workloads fast. This is deep, low-level work (performance counters, memory bandwidth, warp-level scheduling) applied to the specific shapes and patterns our models actually use.\r\nWe hire kernel engineers because the gap between \"this works\" and \"this is fast on the hardware we have\" is enormous, and that gap directly bounds what our researchers can try. You'll close that gap.\r\nWHAT YOU'LL DO\r\nWrite and optimize GPU kernels (CUDA, ROCm, Triton, or similar) for training and inference workloads: attention variants, MoE layers, custom activations, communication primitives\r\nProfile real workloads with hardware counters and translate findings into specific kernel-level optimizations\r\nCo-design kernels with the research teams, when the kernel and the algorithm need to change together, you participate in both\r\nIntegrate optimized kernels into our training and serving stacks; benchmark before and after; verify the win is real end-to-end\r\nMaintain kernel quality over time as hardware, frameworks, and workloads shift underneath\r\nSpread kernel-level fluency across the team; we want this expertise shared, not siloed\r\nWHAT WE'RE LOOKING FOR\r\n4+ years writing performant GPU kernels (CUDA, ROCm, Triton, or production-grade equivalent)\r\nHardware-level fluency: memory hierarchy, occupancy, register pressure, tensor cores, warp scheduling\r\nProfiling fluency (Nsight, ncu, or comparable tools) and the discipline to measure before changing\r\nTrack record of shipping kernel-level optimizations that moved a measurable metric in a real system\r\nStrong systems expertise: you understand how kernels live inside larger frameworks and how integration choices affect end-to-end performance\r\nComfortable reading framework-level Python and C++ around your kernels\r\nNICE TO HAVE\r\nOpen-source contributions to kernel libraries, compilers, or ML frameworks\r\nExperience with multiple accelerator architectures (different GPU families, TPUs, custom ASICs), preferably AMD GPUs\r\nFamiliarity with collective communication primitives (NCCL or equivalent)\r\nCompiler or runtime background\r\nTHIS ROLE IS PROBABLY NOT FOR YOU IF\r\nYou haven't gotten your hands dirty at the kernel level: this isn't a higher-level systems role rebranded\r\nYou want to stay narrowly in one library; we expect breadth across the kernel surface our models actually use\r\nPerformance work without measurable end-to-end impact frustrates you\r\nJ-18808-Ljbffr","datePosted":"2026-06-25T00:42:32.885Z","dateModified":"2026-06-25T00:42:32.885Z","hiringOrganization":{"@type":"Organization","name":"Makermakerai","sameAs":"https://jobsearcher.com"},"jobLocation":{"@type":"Place","address":{"@type":"PostalAddress","addressLocality":"Millbrae","addressRegion":"CA","addressCountry":"US"}},"identifier":{"@type":"PropertyValue","name":"JobSearcher","value":"5214ff3a92227df3211d3814"},"url":"https://jobsearcher.com/jobs/5214ff3a92227df3211d3814"}}