GPU Compute & MLIR Compiler Engineer — AI Workloads

Role Overview

We are looking for a highly skilled GPU compute, MLIR compiler, and kernel optimization engineer with deep expertise in GPU compute, MLIR-based code generation, and end-to-end performance optimization for AI workloads. In this role, you will design, optimize, and deploy high-performance GPU compute kernels, build and extend MLIR compiler backends, and collaborate closely with ML, runtime, and hardware teams to push the limits of performance on modern GPU architectures.

Responsibilities

• Develop and optimize GPU compute kernels targeting OpenCL and Vulkan compute backends for high-throughput AI/ML workloads.
• Design, build, and extend MLIR dialects across multiple abstraction levels including frontend, graph-level, tensor IR, and runtime/low-level dialects.
• Implement and maintain MLIR-based compiler passes and transformations such as tiling, fusion, bufferization, and vectorization.
• Conduct profiling and bottleneck analysis of compiled kernels using GPU counters and vendor-specific profilers.
• Build and maintain GPU runtime infrastructure for both OpenCL and Vulkan, including memory management and command buffer orchestration.
• Develop and extend code generation pipelines for automatic lowering from tensor IR to efficient kernels.
• Implement performance-critical schedules including tiling, loop fusion, and parallelism within MLIR-based backends.
• Collaborate with framework teams to optimize model lowering for computer vision and LLM workloads.
• Design and implement robust compiler and runtime components using modern C/C++.

Requirements

• Strong hands-on experience with MLIR framework, including custom dialects and lowering pipelines.
• Deep expertise across MLIR abstraction levels (TOSA, StableHLO, Linalg, Tensor, Vector, MemRef, SCF, GPU, and LLVM).
• Strong hands-on experience in OpenCL programming and runtime management.
• Solid understanding of Vulkan compute programming and runtime internals.
• Strong understanding of GPU architecture, memory hierarchies, and async compute.
• Proficiency in C/C++ for system-level development.
• Experience with kernel profiling and bottleneck analysis on GPU platforms.
• Strong background in machine learning fundamentals (CV and LLM workloads).
• Bachelor's degree with 4+ years, Master's with 3+ years, or PhD with 2+ years of relevant Systems Engineering experience.

Nice to Have

• Hands-on experience with IREE, TVM, XLA, or LLVM.
• Familiarity with IREE's compiler and runtime architecture.
• Experience contributing to open-source MLIR or IREE projects.
• Knowledge of quantization and mixed-precision inference.
• Exposure to multi-target compilation (CPU, GPU, NPU).
• Familiarity with cross-vendor GPU profiling tools like ARM Streamline, Qualcomm Snapdragon Profiler, or Intel VTune.

Skills

• MLIR
• C/C++
• OpenCL
• Vulkan
• GPU Architecture