Research Scientist, HW/SW Co-Design (PhD)

Job Title

Research Scientist, HW/SW Co-Design (PhD)

Role Summary

Join the AI and Systems CoDesign Group to research and develop hardware-aware techniques that improve performance, efficiency, and programmability for Meta's custom AI accelerators. The role spans workload analysis, model enablement, performance modeling, tooling, and driving prototypes toward production with partner teams.

Work covers model, runtime, compiler, and hardware co-design to optimize generative AI and recommendation workloads at datacenter scale.

Experience Level

Mid-level. PhD in a relevant technical field is expected; candidates with a Masters degree plus 3+ years of relevant industry experience are also considered.

Responsibilities

Contribute research and engineering across hardware and software to optimize AI workloads on Meta's custom accelerators.

• Lead hardware-software co-design for Meta's AI silicon, focusing on programmability, performance, and power efficiency.
• Integrate new silicon and system technologies into the accelerator roadmap based on workload analysis and future model requirements.
• Build system performance models and simulators to evaluate datacenter infrastructure options.
• Co-optimize deep learning kernels and primitives with hardware architects and compiler teams for maximum efficiency.
• Influence architecture and roadmap decisions for custom AI accelerators.
• Architect and implement frameworks and tooling for comparative analyses across system architectures.
• Lead cross-functional initiatives across engineering organizations to achieve high-impact technical milestones.
• Publish research in peer-reviewed conferences and communicate results to technical and non-technical stakeholders.

Requirements

Must-have technical skills and experience.

• Proven research experience in hardware-aware model enablement, performance modeling of AI systems, or accelerator/silicon architecture analysis.
• Hands-on experience with end-to-end AI hardware architecture or on-device mapping algorithm development, including optimizations for performance, power, and area (PPA).
• Theoretical and practical experience with AI models (e.g., CNNs, Transformers, LLMs, Diffusion Models).
• Experience in system-level performance analysis, profiling, and benchmarking of AI workloads.
• Proficiency in Python and experience with at least one major AI framework.
• Track record of publishing in peer-reviewed venues and effectively communicating technical results.

Nice-to-have:

• Experience deploying AI agents and techniques for improved efficiency.
• Experience co-designing hardware/software interfaces or influencing custom silicon architecture.
• Experience with state-of-the-art performance modeling infrastructure and AI model performance projections.
• Familiarity with low-level programming for specialized hardware (CUDA, HIP, Triton) or hardware description languages (HDL).

Education Requirements

PhD in Computer Science, Electrical Engineering, Applied Mathematics, or a related technical field preferred. Alternatively, a Masters degree plus 3+ years of relevant industry experience is acceptable. The posting also references Bachelor's degrees and allows equivalent practical experience. Degree completion is required prior to joining.

About the Company

Company: Meta Platforms

Headquarters: Menlo Park, California, United States

American technology company that develops social networking products (Facebook, Instagram, WhatsApp) and invests in virtual/augmented reality hardware and software through Reality Labs, focusing on connectivity, advertising, and immersive computing experiences.