About the RoleThis is a Research Engineer role focused on building synthetic data pipelines for AI agent training, sitting within a ~15-person engineering team of Olympiad medalists and published researchers. You'll design generation methods, validation systems, and quality metrics that directly expand model capabilities - work that sits at the frontier of RL-based AI alignment.
What You'll Do- Build end-to-end synthetic data pipelines that transform domain-specific workflows into structured, challenging training tasks for AI agents.
- Collaborate with subject-matter experts to develop synthetic tasks spanning professional and technical domains.
- Design task generation methods that produce diverse, realistic, and learnable training examples.
- Build tooling to mutate, validate, and iteratively improve synthetic task quality.
- Analyze model and agent performance on synthetic tasks to understand learning outcomes and failure modes.
- Develop metrics to quantify synthetic task diversity, realism, learnability, and overall quality.
What We're Looking For- 2-4 years of experience in software engineering, ML engineering, or AI research roles delivering data pipelines, ML infrastructure, or synthetic data systems.
- Hands-on experience applying synthetic data research methods to build end-to-end data generation pipelines for AI/ML applications.
- Proficiency in Python and experience developing in Linux environments using containerization tools such as Docker.
- Demonstrated understanding of synthetic data quality criteria and evaluation metrics - diversity, realism, learnability - and their inherent limitations.
- Experience designing, implementing, or maintaining evaluation frameworks, benchmarks, or testing environments for AI agents or large language models.
- Experience building automated systems to generate, validate, mutate, or process structured datasets at scale.
- Track record of independently owning and delivering technical projects end-to-end with minimal predefined requirements.
- Ability to detect edge cases, inconsistencies, and quality issues in synthetic or algorithmically generated datasets.
- Comfort operating in unstructured, early-stage environments and reasoning from first principles.
- Strong communication skills for effective remote collaboration across time zones.
- Nice to have: Familiarity with reinforcement learning paradigms, agentic AI workflows, or LLM post-training pipelines.
Compensation & BenefitsSalary range:
$150,000 - $250,000 USD annually. Visa sponsorship is available.
LocationOn-site in
San Francisco, CA, United States. Singapore-based candidates are also considered.
