Machine Learning Engineer (Model Bring-Up)

Own the end-to-end bring-up of machine learning models for efficient execution on AI accelerator hardware, including validation and performance optimization.

Develop and extend MLIR-based lowering paths, dialects, and transformation passes to map models to hardware-specific execution.

Diagnose performance bottlenecks and collaborate with hardware, compiler, kernel, and runtime teams to ensure correctness and optimize inference latency, throughput, and memory efficiency.

Strong programming skills in C++ and Python.

Experience in bringing up and debugging ML models using PyTorch or similar frameworks.

Practical knowledge of MLIR including dialects, passes, and lowering pipelines.

Understanding of compiler fundamentals, transformer architectures, and profiling/optimizing workloads on AI accelerators.

Engineer with combined expertise in model architecture, compiler infrastructure (MLIR), and systems engineering for AI hardware acceleration.

Experienced in low-level optimization and debugging across model code, compiler-generated code, kernels, and runtime.

Familiar with advanced inference techniques for large language models, precision tradeoffs, and distributed execution concepts.