Senior ML Compiler Architect for LLM Accelerators

Position Overview We are seeking a talented ML Compiler Engineer to join our engineering team and lead the development of a compiler for a novel LLM accelerator architecture. This role focuses on building software systems that bridge high-level AI workloads with custom hybrid optical-electronic compute hardware, enabling breakthrough performance. Key Responsibilities Design and implement toolchains for a custom LLM accelerator architecture Develop optimization strategies that map software algorithms efficiently to hardware implementations Build custom compiler components, including IR dialects, graph transformations, and lowering passes Optimize computational graphs and memory access patterns for specialized hardware Integrate with existing ML frameworks (e.g., PyTorch, JAX, Triton) Develop and maintain testing infrastructure to ensure compiler correctness and performance Qualifications Bachelor’s degree in Computer Science, Computer Engineering, or a related field 10+ years of industry experience 5+ years of experience in systems programming or compiler development Expert-level proficiency in Python and C Experience working with hardware compilers Familiarity with large language model architectures and their computational requirements Hands-on experience with compiler frameworks and optimization techniques Strong understanding of computer architecture, memory hierarchies, and parallel computing Experience with AI/ML accelerators (GPUs, TPUs, FPGAs) and their programming models Preferred Skills Master’s degree in Computer Science, Computer Engineering, or a related field Strong background in graph theory and compiler-based graph transformations (MLIR experience is a plus) Experience working with abstract syntax trees (parsing, analysis, transformation) Experience debugging and instrumenting parallel systems Familiarity with structured, human-supervised AI or agentic coding workflows Experience with LLM quantization and model optimization techniques Background in high-performance computing and low-latency system design Familiarity with deep learning frameworks and neural network optimization Technical Skills Programming Languages: Python, C (essential), Assembly Compiler Frameworks: LLVM, MLIR, GCC, custom backend development Graph Theory: Graph algorithms, DAG optimization, rewriting systems AST Processing: Parsing, analysis, and transformation Testing & QA: pytest, GoogleTest, static analysis tools CI/CD: Jenkins, GitHub Actions, GitLab CI LLM Technologies: Transformer architectures, attention mechanisms, quantization Development Tools: CMake, Git, Docker Parallel Tools: Profilers, debuggers, instrumentation tools Technical Environment Languages: Python and C (primary), Assembly for low-level optimization Compiler Tools: LLVM, MLIR, GCC, custom compiler backends Testing: Automated test suites and continuous integration pipelines Frameworks: PyTorch, JAX, Triton, and custom inference engines Focus Areas: Compiler backend development, optimization passes, and hardware-software co-design #J-18808-Ljbffr