About the company
Our client is an artificial intelligence company.
The role
Raydar is recruiting for this role on behalf of our client. Own the full cycle of improving machine learning models after initial training, from setting up compute to preparing the datasets that drive the work. The role blends applied research with hands-on engineering and calls for sound judgment about which tasks and simulated environments are realistic and worth building.
What you'll do
- Take machine learning models through further training stages end to end, from setting up compute infrastructure to preparing the datasets that feed them.
- Design long-horizon tasks and simulated environments across a range of scientific and technical domains.
- Build automated, self-improving workflows so that output grows faster than headcount.
- Apply and sharpen your judgment about which tasks and environments are realistic and valuable.
- Treat research and dataset production as one integrated function.
Requirements
What we're looking for
- Demonstrated depth in ML or AI research, such as reinforcement learning, evaluations or interpretability.
- Strong software engineering ability alongside that research depth.
- Hands-on experience refining models after initial training and producing well-curated datasets for that purpose.
- Good judgment for what makes a task or environment realistic and valuable, or the research strength to develop it quickly.
- Quick code comprehension and problem-solving, including PyTorch and debugging ML pipelines.
- Genuine, self-driven interest shown through publications, side projects, open-source work or writing.
- Roughly 3 to 7 years of relevant experience.
Bonus points
- Domain expertise in a specific field such as neuroscience, physics, chemistry or systems, paired with an ML and engineering core.
- Background at a leading AI lab or a data-focused technology company.
Benefits
Compensation and benefits
- Base salary: USD 250,000 to 300,000 per year
Location and work model
- San Francisco, CA, United States
- On-site
- Full-time