MLIR/LLVM Compiler Engineer for AI Accelerators

Oho Group is seeking a hands-on compiler engineer to join a well-funded startup building a new ML compiler stack for high-performance accelerators. You will work in the middle of the compiler, transforming models into efficient programs for a new accelerator architecture.

You will own optimizations that affect workload performance, design graph and IR-level optimizations, develop passes, and lower workloads through MLIR-based pipelines while collaborating with architecture and runtime teams.