AI Hardware Performance Engineer – SoC & Simulator

River AI Inc.; Palo Alto, Austin (CA, TX); On-site; USD 200,000 - 420,000; Full time

Benefits offered by this job

Health benefits

Unlimited PTO

Relocation assistance

Job summary

River AI is seeking a hardware performance engineer to architect and model high-performance silicon, developing simulators to predict AI accelerator and SoC behavior for real-world models. You will own performance models, ISA/kernel optimization, and FPGA emulators for software development.

You will collaborate with compiler, IR, and RTL teams across the stack, validating models against RTL and emulators, and identifying bottlenecks from ILP to bandwidth limits within a fast-moving AI hardware

Qualifications

• Bachelor’s degree in Electrical Engineering or Computer Engineering or Computer Science.
• 5+ years of industry experience with advanced process nodes (7nm or below).
• Proficiency in C/C++ or SystemC.
• Experience with compilers (LLVM/GCC/XLA) and high-performance kernels (CUDA/Triton).
• Expert knowledge in Computer Architecture of at least one chip style (SoCs/CPUs/GPUs/AI accelerators).
• Experience profiling hardware with performance counters, hardware profilers, and trace analysis tools.

Responsibilities

• Design and implement high-performance hardware models using C++ and/or SystemC.
• Perform micro-architectural exploration to assess changes and impacts on IPC and execution time.
• Profile AI kernels and software stacks to generate representative traces for stress tests.
• Co-design with compiler and kernel teams to optimize software mapping to hardware.
• Validate performance models against RTL and pre-silicon emulators for accuracy.
• Identify and quantify system bottlenecks from ILP to bandwidth utilization.

Skills

C/C++

SystemC

LLVM/GCC/XLA

Computer Architecture

Performance profiling

Collaboration

RTL/Emulator

Education

Bachelor’s degree in Electrical/Computer Engineering or Computer Science

Tools

QEMU models

LLVM

GCC

CUDA/Triton

Job description

River AI is seeking a hardware performance engineer to architect and model high-performance silicon, developing simulators to predict AI accelerator and SoC behavior for real-world models. You will own performance models, ISA/kernel optimization, and FPGA emulators for software development.

You will collaborate with compiler, IR, and RTL teams across the stack, validating models against RTL and emulators, and identifying bottlenecks from ILP to bandwidth limits within a fast-moving AI hardware