Research Engineer
About the Company: We are on a mission to build safe AGI to accelerate humanity’s progress on the world’s most important problems. Our belief is that automating research and code generation can improve models and solve alignment issues more reliably than humans alone. Our approach utilizes frontier-scale pre-training, domain-specific RL, ultra-long context, and test-time compute to achieve this goal.
About the Role: As a Research Engineer, you will be involved in training, evaluating, and serving large AI models and exploring new test-time compute techniques. You will also build internet-scale datasets and prototype new research and product ideas. This role is essential for optimizing inference throughput for novel model architectures and training trillion-parameter models on large GPU clusters.
What We Can Offer You:
- Significant equity as part of the total compensation
- 401(k) plan with 6% salary matching
- Generous health, dental, and vision insurance for you and your dependents
- Unlimited paid time off
- Option to work in-person in San Francisco or remotely
- Visa sponsorship and relocation stipend available
- A small, fast-paced, highly focused team environment
Key Responsibilities:
- Optimize inference throughput for novel model architectures
- Train trillion-parameter models on large GPU clusters
- Curate post-training datasets to enhance targeted capabilities
- Build out internet-scale data pipelines and crawlers
- Design, prototype, and optimize new model architectures
- Contribute to research in long-context, test-time compute, RL, and more
Relevant Keywords:
Research Engineer, AGI, deep learning, large distributed systems, GPU clusters, internet-scale datasets, frontier-scale pre-training, domain-specific RL, ultra-long context, test-time compute.
