SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Dynamo AI builds a platform for enterprises to deploy and manage AI applications in production, with products covering risk assessment, evaluation, runtime guardrails, and observability for generative AI and agentic systems. The company serves regulated industries including financial services, insurance, healthcare, and defense.
As a Product Analyst, you will own the evidence generation for product decisions. Working closely with Product Managers who own specific business problems—such as getting AI agents approved for production, reducing compliance team effort on guardrails, or lowering guardrail runtime costs—you will take a scoped question and own the answer end-to-end. This includes scoping the question, designing experiments or prototypes, defining success metrics, running analysis, and delivering actionable recommendations.
You may work on diverse problems: designing and curating data for custom guardrail training while maintaining labeling consistency; benchmarking detection against new attack techniques and languages; assessing whether synthetic evaluation and red-teaming data is trustworthy for customer approvers; comparing guardrail and orchestration configurations on accuracy, latency, and cost; prototyping alert aggregation and prioritization systems; building and testing new workflows for legal/compliance review or customer onboarding; and creating small custom tools with AI to validate hypotheses. You will measure performance using standard classification metrics (FNR, FPR, precision, recall) and report findings clearly.
This role offers a growth path toward Product Manager or senior technical product roles. You will start on well-scoped questions and progress to more ambiguous problems, building intuition for SLM training, guardrail orchestration, agent security, and how controls fit into full AI applications.
REQUIREMENTS:
- Degree with strong quantitative or analytical content (physics, information/data theory, business, or similar backgrounds valued alongside solid data analysis experience)
- Clear, demonstrated framework for thinking through problems
- Good instincts for data, ML concepts, and what makes experiments and metrics trustworthy
- Python and pandas proficiency for data work
- Interest in AI security or safety, and adversarial thinking
- Attention to detail and clear writing; ability to turn results into actionable PM recommendations
NICE TO HAVE:
- ML coursework, or coursework/projects in statistics or NLP
- Reading or working proficiency in Japanese, Chinese, or a European language
- SQL
- Habit of building small tools or scripts (including with AI assistants) to answer questions
- Exposure to LLMs, AI agents, security tooling, experimental design, or annotation work
