Senior Compiler Engineer

We are building an MLIR-based compiler and inference stack for custom AI silicon used in telecommunications baseband processing. The compiler takes ML models from StableHLO through orchestration, tiling, memory planning, and code generation onto a multi-core processor architecture with custom accelerators. We already have a working proof point. Real workloads are running on custom silicon at competitive perf-per-watt, and we are building the general-purpose compiler stack around it. This is a small team building something real, not a research project. We are hiring a Staff or Senior Compiler Engineer to work across the compiler stack. You will design and implement MLIR passes, debug IR transformations, optimize for real hardware constraints, and contribute wherever the hardest problems are, whether that is model import, scheduling, memory allocation, or code generation. You will work directly with hardware architects, runtime developers, and AI researchers, and you will help shape the compiler as it matures.

Responsibilities

Design and implement MLIR transformation passes, dialect conversions, and lowering pipelines; Work on tiling, scheduling, vectorization, and memory-planning problems specific to the target hardware; Debug numerical correctness issues across compiler transformations; Develop and maintain compiler testing and validation infrastructure; Contribute to hardware-software co-design discussions as the compiler matures; Profile, analyze, and optimize compiled workloads on silicon; Own the design and implementation of significant compiler subsystems

Requirements

Strong C++ development experience; Hands-on experience with MLIR, LLVM, or another production compiler framework; Experience implementing compiler passes or transformations at production quality; Understanding of intermediate representations and compiler pipelines; Ability to work across abstraction layers, from ML operator semantics to hardware constraints; Track record of owning technical work end-to-end, from design through delivery

Nice to have

StableHLO, XLA, JAX, PyTorch, or ONNX model lowering; Custom MLIR dialect development; DSP, VLIW, or AI accelerator compilation; Static scheduling or software-managed memory; Tiling, loop transformations, or data-layout optimization; Quantization or mixed-precision lowering; Bare-metal or embedded runtime integration

Benefits

Three medical plan options and a dental plan option; Company credits for medical and dental premiums; 401(k) Plan with automatic 3% company contribution and matching contributions; Basic life insurance and basic accidental death and dismemberment coverage; Short-term and long-term disability coverage; Stock Purchase Plan; Minimum of 15 days of accrued vacation; Up to 3 personal days per year; 11 annual holidays; 8 hours of volunteer time; 80 hours of sick time annually; Up to 16 weeks of paid maternity leave; 6 weeks of parental or adoption leave at 100% pay; Financial wellness programs; Educational assistance; Matching gifts; Recognition programs