What are the responsibilities and job description for the Principal Compiler Engineer (ML) position at Oho Group?
About the role:
I'm working with a well-funded startup building a next-generation AI infrastructure platform from the ground up. Their team is developing a modern ML compiler stack that takes models from PyTorch through Triton down to CUDA and machine IR, targeting both today's GPUs and tomorrow's custom AI accelerators.
This is an opportunity to work at the intersection of compilers, machine learning systems, GPUs, and hardware/software co-design, helping define how AI workloads are compiled and executed on cutting-edge hardware.
You'll be:
- Lead the development and optimization of a modern ML compiler stack
- Lowering PyTorch and Triton workloads into efficient machine IR
- Building compiler infrastructure for GPUs (CUDA/ROCm) and emerging AI accelerators
- Working closely with hardware and systems engineers to optimize performance across the stack
- Mentor junior engineers and serve as team lead
Looking for:
- Strong compiler fundamentals and performance optimization experience
- Experience with LLVM, MLIR, or modern compiler infrastructure
- Knowledge of PyTorch, Triton, and low-level code generation
- Background in compilers, ML systems, or high-performance computing
- GPU architecture or hardware/software co-design experience is a strong plus