Job Title

Join to apply for the ML Compiler Engineer - Frontend, Staff role at d-Matrix.At d-Matrix, we focus on unleashing the potential of generative AI to transform technology. We are leaders in software and hardware innovation, emphasizing a culture of respect, collaboration, humility, and direct communication. Our inclusive team values diverse perspectives to develop better solutions. If you are passionate about tackling challenges and driven by execution, join us to shape the endless possibilities of AI.Location: Remote/Hybrid (onsite at our Toronto, Ontario, Canada office 2 days/week)Role: Senior Staff ML Compiler EngineerWhat You Will DoThe d-Matrix compiler team seeks exceptional candidates to develop the front-end of our machine learning compiler. Responsibilities include designing, optimizing, and lowering high-level ML representations to intermediate forms suitable for further compilation. Key contributions include:High-level IR transformations (graph optimization, operator fusion, canonicalization)Dialect and IR design for ML frameworksLowering ML models from frameworks like PyTorch, TensorFlow, ONNX to compiler IRs such as MLIR and LLVMPerformance optimization for compute graphs, including operator specialization, fusion, and memory layout transformationsModel partitioning techniques, such as graph-based parallelism, automatic partitioning with GSPMD, and placement-aware optimizations for distributed hardwareThe successful candidate will collaborate with ML framework developers, hardware architects, and performance engineers to ensure efficient model execution.What You Will BringBachelors degree in computer science or related field + 10+ years of industry experience (or MS + 8+ years, PhD + 3+ years)Proficiency in modern C++ (C++14/17/20) and compiler developmentExperience with LLVM, MLIR, or similar compiler infrastructuresExperience with ML frameworks like PyTorch, TensorFlow, ONNXUnderstanding of graph-level optimizations and IR transformations in ML compilersExperience with model partitioning strategies such as GSPMD, sharding, distributed executionPreferred QualificationsAlgorithm design from concept to implementationExperience with open-source ML compiler projects (Torch-MLIR, IREE, XLA, TVM)Knowledge of automatic differentiation, shape inference, type propagationExperience optimizing distributed execution on accelerators (GPUs, TPUs, custom hardware)Passion for working in a fast-paced startup environmentEqual Opportunity Employment Policyd-Matrix is committed to fostering an inclusive environment and is an equal opportunity employer. We value humility, kindness, dedication, and a willingness to learn and embrace challenges. We do not accept external agency resumes; please apply directly through our official channels.Additional DetailsSeniority level: Mid-Senior levelEmployment type: Full-timeJob function: Engineering and ITIndustry: Semiconductor Manufacturing #J-18808-Ljbffr