Compiler Engineer (AI)

TL;DR

Compiler Engineer (AI): Building compiler pipelines for large-scale AI inference workloads with an accent on multi-level IR design, lowering strategies, and heterogeneous hardware optimization. Focus on improving execution latency, throughput, and memory efficiency through advanced compiler infrastructure and kernel orchestration.

Location: Must be based in the United States (San Francisco, CA preferred).

Salary: $140,000–$325,000

Company

An AI infrastructure company building next-generation systems for large-scale AI workloads and heterogeneous hardware execution.

What you will do

• Build compiler pipelines for large-scale AI inference workloads.
• Design and evolve multi-level IRs across graph, tensor, and kernel representations.
• Implement lowering and partitioning strategies across heterogeneous hardware.
• Optimize execution for latency, throughput, and memory efficiency.
• Improve scheduling, kernel orchestration, and runtime behavior.
• Collaborate with runtime, systems, and kernel engineers to drive end-to-end performance.

Requirements

• Must be based in the United States.
• Strong systems and performance engineering fundamentals.
• Experience building compiler systems or compiler-adjacent infrastructure.
• Deep understanding of execution behavior, memory systems, and hardware efficiency.
• Experience with MLIR, LLVM, XLA, TVM, Triton, or similar infrastructure.
• Strong C++ and/or Python engineering skills.

Culture & Benefits

• Work on cutting-edge compiler infrastructure for modern AI systems.
• Solve complex problems across inference, runtimes, and scheduling.
• Join a small, deeply technical engineering team.
• Significant ownership and technical impact from day one.