Anyone AI is recruiting experienced GPU Kernel Engineers for a specialized project focused on reviewing, debugging, and evaluating high-performance compute kernels used in AI workloads.We're looking for engineers with hands-on experience writing and optimizing kernels across frameworks such as CUDA, Triton, NKI, or Pallas, with a strong understanding of numerical correctness, GPU performance, memory optimization, and benchmarking.What You'll Work OnYou'll work with GPU and accelerator kernel tasks involving:Kernel implementation and debuggingCUDA and Triton optimizationTranslation between kernel frameworksHardware migrationOperator fusionPerformance profiling and benchmarkingNumerical correctness verificationCompilation and runtime debuggingMemory hierarchy optimizationKernel-level AI workload performanceYou'll assess whether implementations are technically correct, efficiently designed, reproducible, and appropriately optimized for the target hardware.What We're Looking For3+ years of hands-on experience developing, optimizing, or debugging GPU or accelerator kernelsStrong experience with at least two of the following:CUDATritonNKI / AWS NeuronPallas / JAXStrong understanding of GPU performance optimizationExperience with kernel profiling tools such as Nsight, NCU, roofline analysis, or framework-native profilersUnderstanding of:Memory bandwidthCompute throughputGPU occupancyShared memoryRegister pressureMemory coalescingBank conflictsStrong understanding of floating-point numerical correctness and tolerance thresholdsExperience debugging kernel compilation and runtime issuesAbility to distinguish software defects, environment problems,



and genuine optimization challengesRelevant ExperienceCandidates should have experience with several of the following types of work:Writing kernels from technical specificationsTranslating kernels between CUDA, Triton, or other frameworksMigrating kernels across hardware platformsDebugging incorrect kernel implementationsOptimizing kernel performanceFusing multiple operations into optimized kernelsNice to HaveExperience across both NVIDIA GPU and custom accelerator ecosystemsExperience with AWS Trainium, TPU, JAX, or other acceleratorsCompiler engineering experienceFamiliarity with MLIR, XLA, or intermediate representation loweringContributions to GPU or ML kernel librariesExperience with cuBLAS, cuDNN, Triton community kernels, or JAX/XLA custom callsExperience with AI model evaluation, RLHF, or technical benchmark developmentWhat You'll Be Responsible ForReviewing GPU and accelerator kernel implementations for correctnessComparing outputs against reference implementationsEvaluating numerical tolerance thresholdsReviewing kernel benchmarks and determining whether comparisons are fairIdentifying performance bottlenecks and optimization opportunitiesAssessing whether performance targets are realistic given hardware limitsReviewing kernel translations and hardware migrationsIdentifying compilation, driver, memory, shape, and runtime issuesDetermining whether technical tasks are genuinely difficult or incorrectly configuredProviding clear, actionable technical feedbackEngagementWork Type: RemoteEngagement: Part-time, project-based consultingFocus: GPU kernels, performance engineering, debugging, and technical evaluationThis role is idóneo for engineers who enjoy working close to the hardware, optimizing GPU workloads, debugging low-level performance issues, and pushing AI compute systems toward their performance limits.
#J-*****-Ljbffr

📌 Gpu Kernel Engineer – Cuda, Triton & Accelerator Performance (La Pampa)
🏢 Anyone AI
📍 La Pampa