Compliant-Ready AI for the Enterprise. Backed by Y Combinator.
About the role
PMs run several workstreams at once. As an analyst, you take individual complex items inside those streams and own their design and specification: scope the problem, design the experiment, define how success is measured, run the analysis, and come back with a recommendation. When building is faster than waiting, that includes standing up a quick custom tool with AI to test an idea. You won't write training or infrastructure code; the platform handles that. What you bring is experimental design and analytical rigor.
What they're looking for
- A degree with strong quantitative or analytical content. We're especially interested in less common backgrounds (physics, information or data theory, business, and similar) alongside solid data-analysis experience
- A clear, demonstrated framework for thinking through problems
- Good instinct for data, ML concepts, and what makes an experiment or metric trustworthy
- Python and pandas for working with data
- Interest in AI security or safety and adversarial thinking
- Attention to detail and clear writing
More about this role
Dynamo AI is building the future of trustworthy AI for the enterprise. Our platform provides real-time guardrails, redteaming, and observability for generative AI systems—ensuring safe, compliant, and reliable AI deployments in regulated industries such as financial services, insurance, DoD, and healthcare.
PMs run several workstreams at once. As an analyst, you take individual complex items inside those streams and own their design and specification: scope the problem, design the experiment, define how success is measured, run the analysis, and come back with a recommendation. When building is faster than waiting, that includes standing up a quick custom tool with AI to test an idea.
You won't write training or infrastructure code; the platform handles that. What you bring is experimental design and analytical rigor.
Guardrail training data : curate and design the data that trains an SLM guardrail (for example, a prompt injection detector), and keep labeling consistent as datasets grow.
Orchestration experiments : test different configurations of guardrails and controls against each other to find what performs best.
Synthetic data quality : assess whether synthetic training or...
Browse similar: AI jobs · AI startup jobs · Startup jobs · San Francisco Bay Area