company

Reinforcement Learning Engineer at Code Metal AI | Remote

Code Metal AIRemote — San Francisco, California, United States
Remote Full-time

Clicking Apply Now takes you to AutoApply where you can tailor your resume and apply.


Unlock Your Potential

Generate Job-Optimized Resume

One Click And Our AI Optimizes Your Resume to Match The Job Description.

Is Your Resume Optimized For This Role?

Find Out If You're Highlighting The Right Skills And Fix What's Missing

Experience Level

Experience

Qualifications

Minimum of 2 years of experience in distributed training, preferably using PyTorch. Strong foundation in reinforcement learning, with recent RLHF experience being highly preferred. Demonstrated ability to construct data curation and quality assurance pipelines. Experience in developing evaluation frameworks. Ideally, familiarity with both data pipeline and orchestration aspects. Eligibility for TS/SCI clearance.

About the job

Join Code Metal AI's elite team, comprised of talents from MIT, OpenAI, and other esteemed organizations, as we lead the charge in pioneering large language models (LLMs) and advanced code generation techniques. Our innovative projects engage with top-tier chip manufacturers, leveraging cutting-edge AI to tackle significant, real-world challenges.

This position serves as a critical link between two essential domains:

Production Responsibilities:

  • Establish and uphold resilient distributed training systems utilizing PyTorch (2+ years of experience required).
  • Design and execute scalable data curation and quality assurance pipelines to ensure high-quality training datasets.
  • Create orchestration tools that streamline complex workflows for large-scale AI model training and evaluation.

Research Responsibilities:

  • Lead the innovation in developing evaluation frameworks and reinforcement learning solutions, emphasizing recent advancements in Reinforcement Learning with Human Feedback (RLHF).
  • Engage with cutting-edge research through open-source contributions and potential publications, focusing on applying RLHF to LLMs, particularly in code generation tasks.

Qualifications:

  • Minimum of 2 years of experience in distributed training, preferably using PyTorch.
  • Strong foundation in reinforcement learning, with recent RLHF experience being highly preferred.
  • Demonstrated ability to construct data curation and quality assurance pipelines.
  • Experience in developing evaluation frameworks.
  • Ideally, familiarity with both data pipeline and orchestration aspects.
  • Eligibility for TS/SCI clearance.

Preferred Qualifications:

  • Contributions to open-source AI or ML initiatives.
  • Published research or experience in relevant fields.
  • Hands-on experience implementing RLHF to LLMs, especially for code generation.
  • Experience in large-scale synthetic data generation.

Benefits:

  • Comprehensive healthcare plan with 100% premium coverage, including medical, dental, and vision.
  • 401k plan with 5% matching contribution.
  • Unlimited Paid Time Off, along with Sick leave and Public Holidays.
  • Flexible hybrid work arrangement.
  • Relocation assistance for eligible employees.

About Code Metal AI

At Code Metal AI, we pride ourselves on assembling a world-class team, drawing talent from prestigious institutions like MIT and OpenAI. Our focus is on groundbreaking advancements in large language models and innovative code generation, collaborating directly with leading chip manufacturers to address impactful, real-world challenges.

Similar jobs

Tailoring 0 resumes

We'll move completed jobs to Ready to Apply automatically.