Research Scientist / Research Engineer

Role Overview

We are a small, driven team focused on creating the next generation of AI training datasets, prioritizing accuracy and verifiability over noisy human-generated or web-sourced data. Our approach integrates formal systems such as physics simulators, scientific databases, formal proof systems, executable tests, and oracle databases to generate data that is inherently correct.

Key Responsibilities

• Develop systems that produce high-quality, verifier-validated training data for large language models, robotics, and scientific AI applications.
• Create data pipelines leveraging formal verification systems to ensure correctness and scalability.
• Conduct experiments assessing the quality of the data and its influence on model performance.
• Work closely with the founding team to define research priorities and focus on impactful initiatives.
• Depending on the role, either implement robust production-quality solutions (Research Engineer) or lead pioneering methodological research (Research Scientist).

Candidate Profile

• Experience varying from 0 to 12 years is welcomed, emphasizing deep analytical thinking and successful development or discovery experience.
• Strong grounding in machine learning, deep learning, or related disciplines, via professional experience, graduate education, or significant self-directed projects.
• Familiarity with large language models, generative modeling, or reinforcement learning is highly advantageous.
• Comfortable working within at least one domain of robotics, scientific computing, formal verification methods, or data pipeline engineering.
• Proficient in software engineering, with ability to prototype rapidly and iterate solutions; mastery of Python is required.
• Research Engineers should have strong skills in designing scalable systems, building data infrastructure, and operationalizing research concepts into dependable pipelines.
• Research Scientists should focus on innovative method development, rigorous experimentation, and demonstrated research output such as publications or equivalent work.
• A deep passion for understanding how data influences AI model capabilities and critical perspectives on prevailing industry paradigms.

Additional Terms

Applicants must be legally authorized to work in the United States without visa sponsorship, as no sponsorship is provided for this position.

Compensation and Benefits

• An annual salary between 100000 and 300000 USD, reflecting the range of roles and experience levels.
• Opportunity to receive early-stage equity in the company.
• A work environment characterized by high autonomy and ownership at a seed-stage startup backed by reputable investors.

Location and Work Arrangement

• This is a fully onsite position based in San Francisco, California.
• Remote work options are not available for this role.