Machine Learning Compiler EngineerFor a high growth Series A Deep Tech Company with over $50 million infunding, we’re seeking a Machine Learning Compiler Engineer to jointheir growing team.Responsibilities:Lower deep learning graphs – from commonframeworks (PyTorch, Tensorflow, Keras, etc) down to an IR representationfor training – with particular focus on ensuring reproducibilityWrite novel algorithms – for transformingintermediate representations of compute graphs between different operatorrepresentations.Ownership – of two of the following compilerareas:Front-end: Integrate common Deep Learning Frameworkswith our internal IR, and implement transformation passes in ONNX to adaptIR for middle-end consumption.Middle-end: Design compiler passes for training-basedcompute graphs, integrate reproducible Deep Learning kernels into the codegeneration stage, and debug compilation passes and transformations.Back-end: Translate IR from the middle-end to GPUtarget machine code.Required Skills:Fundamental knowledge of traditional compilers (e.g., LLVM, GCC) andgraph traversals necessary for compiler code development.Strong software engineering skills, demonstrated by contributing to anddeploying production-grade code.Understanding of parallel programming, particularly concerningGPUs.Willingness to learn Rust, as it is our company’s defaultprogramming language.Ability to operate with High-Level IR/Clang/LLVM up to middle-endoptimization, and/or Low-Level IR/LLVM targets/target-specificoptimizations, especially GPU-specific optimizations.Highly self-motivated with excellent verbal and written communicationskills.Comfortable working independently in an applied researchenvironment.Bonus Skills:Thorough understanding of computer architectures specialized fortraining neural network graphs (e.g., Intel Xeon CPU, GPUs, TPUs, customaccelerators).Experience in systems-level programming with Rust.Contributions to open-source Compiler Stacks.In-depth knowledge of compilation in relation to High-PerformanceComputer architectures (CPU, GPU, custom accelerator, or a heterogeneoussystem).Strong foundation in CPU and GPU architectures, numeric libraries, andmodular software design.Understanding of recent architecture trends and fundamentals of DeepLearning, along with experience with machine learning frameworks and theirinternals (e.g., PyTorch, TensorFlow, scikit-learn, etc.).Exposure to Deep Learning Compiler frameworks like TVM, MLIR,TensorComprehensions, Triton, JAX.Experience in writing and optimizing highly-performant GPUkernels.We are excited to hear from youPlease apply for more details on this wonderful opportunity! Show more Show lessTagged as: remote, remote job, virtual, Virtual Job,virtual position, Work at Home, work from home