Member of Technical Staff - Compiler Engineer (AI)

Location: San Francisco, CA; on-site

Salary: $180,000–$400,000 per year

Company

hirify.global is building a next-generation compute platform for fast and efficient AI inference across heterogeneous hardware.

What you will do

• Build and scale an AI compiler stack based on MLIR and LLVM.
• Design custom dialects, intermediate representations, optimization passes, and lowering paths for CPUs, GPUs, and emerging accelerators.
• Develop high-performance code generation, operator fusion, tiling, layout transformation, memory planning, and execution-planning optimizations.
• Optimize workload partitioning, device-to-device state movement, latency, throughput, utilization, and serving cost.
• Integrate generated, library, and hand-written kernels into performance-driven code-generation strategies.
• Collaborate with kernel, runtime, ML systems, distributed-systems, and infrastructure engineers.

Requirements

• Experience building ML compilers or runtimes with a strong focus on performance optimization.
• Experience with MLIR, LLVM, intermediate representations, compiler passes, lowering, and code-generation pipelines.
• Strong C++ and/or Python skills.
• Strong understanding of SSA, memory systems, scheduling, and hardware efficiency.
• Experience with operator fusion, tiling, layout transformations, memory planning, and performance profiling.
• Familiarity with IREE, XLA, TVM, Triton, or similar ML compiler and accelerator-programming frameworks.

Nice to have

• Experience optimizing ML inference or model-serving workloads.
• Kernel dispatch, runtime-interface development, memory allocators, or automated kernel generation.
• Autotuning, search-based optimization, or contributions to open-source compiler infrastructure.
• Experience with heterogeneous execution, distributed inference, speculative decoding, or prefill/decode disaggregation.

Culture & Benefits

• Work on frontier models, production infrastructure, and emerging accelerator architectures.
• Contribute to a platform supporting multiple heterogeneous compute targets.
• Collaborate across compiler, kernel, runtime, ML systems, distributed-systems, and infrastructure disciplines.