Research Engineering Specialist - Llama, GenAI

Meta, City of Westminster

Research Engineering Specialist - Llama, GenAI

Salary not available. View on company website.

Meta, City of Westminster

  • Full time
  • Permanent
  • Onsite working

Posted today, 31 Oct | Get your application in now to be one of the first to apply.

Closing date: Closing date not specified

job Ref: c0dfcad7ea034cee83e8b55143ca318c

Full Job Description

The Llama, GenAI team is seeking an AI Specialist for a Research Engineering role. This position involves advancing reasoning capabilities in large language models. In this role, you will work with our state-of-the-art Llama, model and implement cutting-edge techniques, including but not limited to Monte Carlo tree search, online reinforcement learning and advanced efficient inference methods with Llama. The focus will range from verifiable reasoning, such as mathematical reasoning, to more complex areas, including legal reasoning, causal analysis and decision-making., 1. Execute research to push forward state-of-the-art reasoning capabilities in large language models.
2. Contribute to code health, research experiments, and organization of research insights, and communicate results effectively.
3. Collaborate with a global team from diverse backgrounds on the entire large language model post-training pipeline.
4. Thrive in a fast-paced environment with ambitious targets.

5. A minimum of a Bachelor's degree in Computer Science, Computer Engineering, Artificial Intelligence, Machine Learning, or a related field.
6. Relevant industry or research experience related to the job, including experience in collaborative team environments.
7. Proficiency in software development and deep learning frameworks, such as PyTorch, and others.
8. Specialised experience in large language model reasoning through post-training, alignment demonstrated by past projects., 9. An advanced degree in machine learning and natural language processing (Master's or PhD) is preferred.
10. Experience in reasoning (math and code and other non verifiable reasoning) space with LLMs.
11. Experience in reinforcement learning and tree search algorithms
12. Experience in various reword modeling in LLM alignment

Relevant jobs