Etched
Inference Software Engineer
$175K–$400K
San Jose · On-site · Full-time · Posted Jun 17, 2025
About the job
Build and optimize software that ports AI models to Etched’s hardware and runs inference across nodes.
What you'll do
- Support model porting and develop programming abstractions and testing capabilities.
- Build and scale the inference runtime, including execution, state management, and error handling.
- Optimize routing and communication, and profile and debug performance and correctness.
What they ask for
- Proficiency in C++ or Rust.
- Understanding of performance-sensitive or complex distributed software systems, such as Linux internals, accelerator architectures, compilers, or high-speed interconnects.
- Familiarity with PyTorch or JAX.
Nice to have
- Experience porting applications to non-standard accelerator hardware or platforms.
- Experience with networking stacks, distributed systems, Transformer/MoE architectures, or SIMD optimization.
Pay
$175K – $400K USD base salary
Tools
C++, Rust, Linux, GPUs, TPUs, NVLink, InfiniBand, PyTorch
AI focus
Port state-of-the-art models and build inference runtime and communication software for Etched’s architecture.
Model-written summary · the original posting is authoritative