AI Research Scientist, SysML

Meta•Sunnyvale, CA

1d•$184,000 - $257,000

About The Position

Reality Labs (RL) focuses on delivering Meta's vision through virtual reality, augmented reality, and wearable devices. Because these products require high compute performance and power efficiency, the Reality Labs Silicon team develops custom silicon and advances computer vision, machine learning, mixed reality, graphics, displays, sensors, and human-body mapping. Its work spans the full stack, from transistors through architecture, firmware, and algorithms. Meta is seeking a Research Scientist to join its research and development teams. The role focuses on AI models, hardware acceleration, and software systems, applying those skills to challenging Reality Labs problems. The primary objective is to develop novel solutions that enable compute- and power-efficient training and on-device inference of vision and language models for augmented reality, virtual reality, and edge-device use cases. The position is available in multiple locations.

Requirements

Bachelor's degree in Computer Science, Computer Engineering, a relevant technical field, or equivalent practical experience.
PhD in Electrical Engineering, Computer Science, or equivalent experience.
Experience developing AI-system infrastructure, AI algorithms, or AI hardware acceleration in C/C++ or Python.

Nice To Haves

Experience working and communicating cross-functionally in a team environment.
Ability to integrate AI tools to optimize or redesign workflows and drive measurable impact, such as efficiency gains or quality improvements.
Experience or knowledge of architecting machine-learning hardware accelerators and systems.
Experience or knowledge of post-training inference for large-scale generative AI or LLM models.
A track record of significant results demonstrated through grants, fellowships, patents, or publications at leading workshops, journals, or conferences such as ICLR, NeurIPS, CVPR, ACL, ICML, MLSys, ISCA, MICRO, DAC, or ASPLOS.
Experience adhering to and implementing responsible, ethical AI practices, such as risk assessment, bias mitigation, and quality and accuracy reviews.
Ongoing AI skill development, such as prompt or context engineering and agent orchestration, and staying current with emerging AI technologies.
Experience evaluating alternative system or algorithm designs by analyzing performance, power, and latency trade-offs to recommend a solution.
Experience with PyTorch, TensorFlow, or similar machine-learning toolsets.
Research and engineering experience demonstrated through an internship, work experience, coding competitions, or widely used contributions to open-source repositories such as GitHub.
Experience or knowledge of on-device algorithm development, including hardware-aware machine-learning models and/or optimizing machine-learning compilers for efficient deployment on AI accelerators.

Responsibilities

Identify and solve multidisciplinary machine-learning acceleration problems across algorithms, model architecture, software, and hardware for AR, VR, and wearable use cases, including problems that require novel approaches.
Invent and develop state-of-the-art model compression, efficiency, and scalability techniques, including numerics, pruning, distillation, speculative decoding, and memory-efficient transformer architectures; benchmark them across diverse network classes.
Integrate proven research into production machine-learning infrastructure for broad organizational adoption.
Post-train and optimize foundation models to deliver Pareto-optimal multimodal LLMs and generative AI models on target hardware under real-time latency, power, and memory constraints.
Use model and algorithm insights to co-design next-generation hardware and system architectures in partnership with research scientists and silicon teams, and drive the algorithms and software integration needed to use these enhancements.
Define use cases, methodologies, and benchmarks for evaluating competing approaches, and influence partners to adopt recommended solutions through data-driven analysis and clear communication of trade-offs.
Stay at the forefront of efficient on-device machine-learning research and contribute to patents and publications in peer-reviewed conferences and journals.