Compiler Engineer (San Mateo)

Compiler EngineerWe're partnered with an AI infrastructure company building next-generation systems for large-scale AI workloads.Their platform is rethinking how AI execution runs across heterogeneous hardware — dynamically partitioning workloads, optimizing execution paths, and orchestrating inference across multiple accelerator types to unlock major gains in performance and efficiency.What You'll Work OnBuilding compiler pipelines for large-scale AI inference workloadsDesigning and evolving multi-level IRs across graph, tensor, and kernel representationsImplementing lowering and partitioning strategies across heterogeneous hardwareOptimizing execution for latency, throughput, and memory efficiencyImproving scheduling, kernel orchestration, and runtime behaviorWorking on compiler infrastructure spanning execution systems, runtime optimization, and memory movementIntegrating new model architectures and inference execution patterns into the stackCollaborating closely with runtime, systems, and kernel engineers to drive end-to-end performance improvementsWhat They're Looking ForStrong systems and performance engineering fundamentalsExperience building compiler systems or compiler-adjacent infrastructureDeep understanding of execution behavior, memory systems, and hardware efficiencyExperience with MLIR, LLVM, XLA, TVM, Triton, or similar compiler/runtime infrastructureFamiliarity with IR transformations, compiler passes, lowering systems, or runtime optimizationExperience working with GPUs, AI accelerators, or heterogeneous compute environmentsStrong C++ and/or Python engineering skillsWhy JoinWork on cutting-edge compiler infrastructure for modern AI systemsSolve hard problems across inference, runtimes, scheduling, and heterogeneous executionBuild foundational infrastructure for next-generation AI workloadsSmall, deeply technical engineering teamSignificant ownership and technical impact from day one