6 months ago
Austin, TX, USA or Sunnyvale, CA, USAStaff+
Base Salary
$240k - $280k/yr
Responsibilities
- Design and implement toolchains for a custom LLM accelerator architecture.
- Develop optimization strategies that bridge software algorithms to hardware implementations.
- Implement compiler components including IR dialects, graph transformations, and lowering passes.
- Optimize computational graphs and memory access patterns for the hardware architecture.
- Integrate the compiler with PyTorch, JAX, Triton, and custom inference engines.
- Build and maintain test infrastructure for compiler correctness and performance.
- Contribute to compiler backend development, optimization passes, and hardware-software co-design.
Requirements
- Master's degree in computer science or a related field.
- At least 5 years of compiler development experience.
- Expert-level proficiency in Python and C.
- Experience with hardware compilers, compiler frameworks, and code optimization techniques.
- Familiarity with large language model architectures and their computational requirements.
- Deep understanding of computer architecture, memory hierarchies, and parallel computing.
- Experience with AI/ML accelerators such as GPUs, TPUs, or FPGAs and their programming models.
- Preferred qualifications include a PhD, 7+ years of industry experience, MLIR experience, graph theory and graph transformation experience, AST processing, parallel-program debugging and instrumentation, LLM quantization, high-performance computing, low-latency system design, and neural network optimization.
- Experience with Python, C, Assembly, LLVM, MLIR, GCC, testing frameworks, CI systems, CMake, Git, and Docker.
Benefits
- 100% coverage of base health plan premiums for employees and dependents, plus HSA contributions.
- Unlimited PTO.
- 401(k) matching and stock option opportunities.
- Dental, vision, life, hospital, critical illness, and accident insurance.
- Personalized benefits options with cash back for declined plans.
- Full-time onsite work in San Jose, CA, or Austin, TX.
