Summary:

Meta is seeking a Visiting Researcher to join our Systems ML - SW/HW Co-Design team to focus on end-to-end sub-8 bit training experimentation on LLaMa modelsAI System SW/HW Co-design team’s mission is to explore, develop and help productize high-performance software and hardware technologies for AI at datacenter scale. We achieve this via concurrent design and optimization of many aspects of the system such as models, algorithms, numerics, performance and AI hardware including compute, networking and storage. In essence, we drive the AI HW roadmap at Meta and ensure our existing and future AI workloads and software are well optimized and suited for the hardware infrastructure.

Required Skills:

Visiting Researcher, Systems ML - SW/HW Co-Design Responsibilities:

Implement the quantization emulation code in the xlformers for LLaMa4 training and be compatible with the model parallelism strategies

Optimize the emulation operator performance to reduce the latency of the end-to-end training.

Add the periodic quantization error logging to signal the training divergence and facilitate debugging.

Compare different quantization strategies on the train loss curves.

Evaluate the post-training sub 8-bit inference accuracy for final validations.

Minimum Qualifications:

Minimum Qualifications:

Currently has, or is in the process of obtaining a Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience. Degree must be completed prior to joining Meta.

Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience.

Specialized experience in one or more of the following machine learning/deep learning domains: Hardware accelerators architecture, GPU architecture, machine learning compilers, or ML systems, AI infrastructure, high performance computing, performance optimizations, or Machine learning frameworks (e.g. PyTorch), numerics and SW/HW co-design

Experience developing AI-System infrastructure or AI algorithms in C/C++ or Python

Must obtain work authorization in the country of employment at the time of hire and maintain ongoing work authorization during employment

Preferred Qualifications:

Preferred Qualifications:

PhD degree in Computer Science, Computer Engineering

Experience with distributed systems or on-device algorithm development

Experience with large language models

Experience collaborating with other teams in a fast-paced environment

Public Compensation:

$56.25/hour to $137,000/year + benefits

Industry: Internet

Equal Opportunity:

Meta is proud to be an Equal Employment Opportunity and Affirmative Action employer. We do not discriminate based upon race, religion, color, national origin, sex (including pregnancy, childbirth, or related medical conditions), sexual orientation, gender, gender identity, gender expression, transgender status, sexual stereotypes, age, status as a protected veteran, status as an individual with a disability, or other applicable legally protected characteristics. We also consider qualified applicants with criminal histories, consistent with applicable federal, state and local law. Meta participates in the E-Verify program in certain locations, as required by law. Please note that Meta may leverage artificial intelligence and machine learning technologies in connection with applications for employment.

Meta is committed to providing reasonable accommodations for candidates with disabilities in our recruiting process. If you need any assistance or accommodations due to a disability, please let us know at [email protected].