Systems ML Tooling Engineer for AI Accelerators

Meta; United States; Remote; USD 120,000 - 180,000; Full time

Job summary

Meta is seeking a Software Engineer to join the MTIA Software Tooling team, building tooling for Meta's in-house AI accelerator. You will design, implement, and maintain developer tools to debug, profile, measure, and monitor AI workloads on MTIA hardware at scale, spanning CPUs, GPUs, and accelerator stacks.

You will collaborate across compiler, runtime, hardware, and ML framework domains to deliver a high-quality developer experience for Meta's custom AI accelerators.

Qualifications

• Bachelor's degree in CS/CE or equivalent completed before joining Meta.
• 2+ years of software engineering experience with systems software, tooling or infrastructure.
• Proficiency in C++ and Python with systems-level programming.
• Experience across multiple layers of a system stack (application, runtime, OS/driver, hardware).
• Track record delivering software from design to production deployment.
• Experience debugging complex software using logs, stack traces, or diagnostic tools.
• Exposure to accelerator ecosystems (GPU/CUDA, TPU, custom ASICs) or heterogeneous computing.
• Ongoing AI skill development and familiarity with emerging AI technologies.
• Experience building developer tools (debuggers, profilers, build systems, CLI tools, monitoring dashboards).
• Contributions to open-source projects demonstrating tooling or system software interest.
• Data-driven Methods to evaluate tooling effectiveness and improvements.
• Familiarity with Linux debugging and profiling tools (gdb, perf, strace, eBPF, valgrind).
• Familiarity with ML frameworks (PyTorch, TensorFlow) or compiler infrastructure (LLVM, MLIR, TVM).
• Experience with distributed systems debugging, profiling, or monitoring at scale.
• Ability to integrate AI tools to optimize workflows and drive impact.

Responsibilities

• Design, build, and maintain developer tools to help engineers debug, profile, measure, and monitor AI workloads on MTIA hardware at scale.

Skills

C++

Python

Systems software

Developer tooling

Infrastructure

Debugging complex software systems

AI tooling awareness

Open-source contributions

Data-driven tooling evaluation

Linux debugging tools

GPU/CUDA / accelerator ecosystems

ML frameworks familiarity

Distributed systems debugging

Cross-functional collaboration

Education

Bachelor's degree in Computer Science or Computer Engineering or related field

Tools

Linux debugging tools (gdb, perf, strace, eBPF, valgrind)

LLVM

MLIR

TVM

PyTorch

TensorFlow

CUDA

Job description

Meta is seeking a Software Engineer to join the MTIA Software Tooling team, building tooling for Meta's in-house AI accelerator. You will design, implement, and maintain developer tools to debug, profile, measure, and monitor AI workloads on MTIA hardware at scale, spanning CPUs, GPUs, and accelerator stacks.

You will collaborate across compiler, runtime, hardware, and ML framework domains to deliver a high-quality developer experience for Meta's custom AI accelerators.