ML Compiler & Codegen Engineer for Heterogeneous Backends

General Diffusion, Inc.; San Francisco, Northern (CA, KY); Hybrid; USD 180,000 - 260,000; Full time

Job summary

General Diffusion, Inc. is seeking a Member of Technical Staff to translate workloads into efficient executables across diverse accelerators. You will shape IR contracts, implement compiler passes, and verify numerical correctness with automated tests.

Ideal candidates bring MLIR/LLVM, C++, Python, or Rust expertise and a track record in performance tuning, cross-target validation, and collaboration with runtime and safety teams.

Qualifications

• Experience shipping compiler passes, lowering pipelines, code generation, or accelerator backends—not solely using compiler frameworks.
• Strong command of MLIR, LLVM, XLA/StableHLO, Triton, TVM, Mojo, or comparable compiler stacks, including IR design and dialect/lowering mechanics.
• Proven performance-engineering practice: profiling and benchmarks to distinguish graph, memory, runtime, and generated-code effects.
• Ability to reason rigorously about compiler correctness, numerical behavior, shape/layout constraints, and target-specific legality across multiple hardware classes.
• Production-quality systems engineering in C++, Rust, Python, or similar, with disciplined debugging, tests, and reproducible performance investigations.

Responsibilities

• Design and evolve target-independent IR contracts, legality rules, and staged lowering boundaries that preserve workload semantics while allowing backend specialization.
• Implement and maintain compiler passes, analyses, and target backends that lower ML workloads into executable programs for heterogeneous accelerators.
• Build generated-code conformance, semantic-equivalence, and numerical-correctness tests against defined references; make unsupported cases and assumptions explicit.
• Profile compiled workloads to identify graph, memory, scheduling, and code-generation bottlenecks, then turn findings into reproducible optimizations and benchmark regressions.
• Define machine-readable backend capability, compilation-cost, and execution-characteristic reports for Runtime & Placement.
• Collaborate with Kernels on compiler-to-kernel interfaces and with Safety & Formal Verification on evidence and hooks for enforcement.

Skills

MLIR

LLVM

XLA/StableHLO

Triton

TVM

C++

Python

Rust

Job description

General Diffusion, Inc. is seeking a Member of Technical Staff to translate workloads into efficient executables across diverse accelerators. You will shape IR contracts, implement compiler passes, and verify numerical correctness with automated tests.

Ideal candidates bring MLIR/LLVM, C++, Python, or Rust expertise and a track record in performance tuning, cross-target validation, and collaboration with runtime and safety teams.