Meta is hiring a Software Engineer, Systems ML - Edge Inference!

On-Device AI is a critical part of virtual and augmented reality systems. In this role, you will work inside Meta Reality Labs to enable and optimize a wide range of state-of-the-art deep learning models (including Vision, Speech, Codec Avatars, LLMs, Gen AI, etc.), on a variety of VR & AR devices (including VR headsets such as Meta Quest, Smartglasses like Ray-Ban Meta, EMG Wristbands, and other upcoming products).

This role is focused on efficient ML inference via use of edge hardware accelerators, including NPUs and DSPs. The position requires a combination of expertise in machine learning and software engineering.

Software Engineer, Systems ML - Edge Inference Responsibilities

• Contribute to the development of machine-learning libraries, intermediate representations, export formats, and analysis tools.
• Profile models to analyze performance and power efficiency of deep learning inference workloads.
• Partner with teams across Reality Labs to ship models in production devices Map ML graphs to machine learning accelerator hardware implementations, model data-flows, create cost-benefit analysis and estimate silicon power and performance.
• Design effective compiler passes and optimizations.
• Implement ML operators using low-level instructions & compiler intrinsics.

Minimum Qualifications

• Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience.
• 3+ years of experience in ML framework development or accelerating deep learning models on hardware architectures.
• Experience developing AI-System infrastructure or AI algorithms in C/C++.

Preferred Qualifications

• Masters or PhD in Computer Science, Computer Engineering.
• 5+ years of experience in ML framework development or accelerating deep learning models on hardware architectures.
• Experience with PyTorch.
• Experience with compiler back-ends.
• Experience with SIMD vector programming.
• Experience with NPUs, DSPs, and ML accelerators.