Senior AI Product Manager, Evals & Annotations

About Arize

AI is rapidly transforming the world. As generative AI reshapes industries, teams need powerful ways to monitor, troubleshoot, and optimize their AI systems. That’s where we come in. Arize AI is the leading AI & Agent Engineering observability and evaluation platform, empowering AI engineers to ship high-performing, reliable agents and applications. From first prototype to production scale, Arize unifies build, test, and run in a single workspace, so teams can ship faster with confidence.

Arize serves a rapidly growing customer base of 150+ leading enterprises and Fortune 500 companies. Customers like Booking.com , Uber, Siemens, and PepsiCo leverage Arize to deliver AI that works.

The Opportunity

Evaluations are central to how AI developers understand, measure, and improve their applications. As AI systems become increasingly complex and agentic, developers need cohesive workflows to define what good looks like, evaluate behavior, incorporate human feedback, and continuously improve their applications.

We're looking for a highly technical, product-minded PM to own Evals & Annotations at Arize. You'll own the end-to-end evaluation lifecycle—from eval creation and execution to human review, analysis, and iteration. This is an opportunity to bring together powerful capabilities into a cohesive, best-in-class developer experience while helping define the future of evaluations in a world of increasingly autonomous and managed agents, including through our integration with Dynatrace.

This role is ideal for someone with strong product and UX opinions who doesn't just define requirements but actively shapes the experience. You'll independently explore workflows, build prototypes, and collaborate closely with Design and Engineering to turn ideas into polished, technically sound products. You should be as comfortable working through an interaction design as you are discussing technical tradeoffs with engineers.

While Arize has been acquired, we continue to operate with the urgency, resourcefulness, and hands-on mentality of an early-stage startup. Product, Engineering, and Design collaborate closely, and our founders remain actively involved in product thinking and design. You'll have real ownership while working directly with our cofounder/CPO in an environment that values initiative, constructive debate, and a willingness to jump in wherever needed.

What You'll Do

• Own Arize's Evals & Annotations product area end-to-end, including eval creation and execution, human review, annotation workflows, and the broader evaluation lifecycle.
• Set the product strategy and roadmap, balancing customer needs, foundational platform investments, UX improvements, and emerging opportunities in AI and agent development.
• Drive a best-in-class evaluation experience. Develop strong opinions about how workflows should work, independently prototype solutions, and partner with Design to turn those ideas into intuitive, polished experiences.
• Shape the future of evaluations as AI development evolves toward autonomous and managed agents, including opportunities within the broader Dynatrace ecosystem.
• Work closely with Engineering on technically complex problems spanning evaluation execution, backend systems, orchestration, and data infrastructure. Contribute to technical discussions, implementation approaches, and tradeoffs.
• Spend significant time with customers—typically multiple conversations each week—to understand how they evaluate and improve AI systems, uncover underlying problems, and validate product direction.
• Collaborate directly with our cofounders and cross-functional teams on product direction, design exploration, and prioritization, bringing independent judgment and turning collaborative thinking into execution.
• Make clear prioritization decisions across competing demands and drive products from problem definition through launch, adoption, and iteration.

What We're Looking For

• Significant product management experience independently owning technically complex developer-facing products. We're flexible on exact years of experience for candidates with exceptional technical depth, product judgment, and ownership.

• Strong understanding of LLM evaluations and AI development workflows, including automated evals, human feedback, experimentation, and how developers iterate on AI applications and agents.

• Experience with developer tools, AI infrastructure, ML platforms, or adjacent technical products, with a solid understanding of backend systems and software architecture.

• Strong technical depth and hands-on engineering ability. We don't expect you to ship large features as an IC engineer, but we do expect you to:

• Debug issues independently and navigate code when needed.

• Put up small fixes or PRs.

• Build prototypes and technical proofs of concept.

• Propose technical implementations and reason through tradeoffs.

• Hold credible technical discussions with engineers.

• Excellent product and UX instincts.You can independently define and prototype core product experiences, not just write requirements, while partnering effectively with designers and design engineers to refine and ship them.

• Strong ownership, initiative, and bias toward action. You're comfortable navigating ambiguity, making tradeoffs, and getting hands-on without waiting for a fully defined roadmap or direction.

• A collaborative, low-ego approach to working with founders, engineers, and designers. You bring strong opinions, engage in constructive debate, and thrive in an environment where responsibilities overlap.

• Experience working closely with technical customers, with strong communication skills and the ability to move comfortably between strategy and execution.

Bonus Points, But Not Required

• Direct experience with LLM evaluation platforms, annotation systems, human-in-the-loop workflows, or agent orchestration.
• Strong backend or platform engineering background, particularly with distributed systems, workflow orchestration, or data infrastructure.
• Hands-on programming experience and comfort working across backend and frontend technologies.
• Experience with LLM-as-a-judge, custom evaluators, experiment tracking, datasets, or feedback-driven improvement workflows.
• Previous engineering experience or substantial hands-on software development experience.
• Early-stage or high-growth startup experience, particularly in founder-led environments.

The estimated annual salary for this role is between $200,000-$250,000, plus a competitive equity package. Actual compensation is determined based upon a variety of job related factors that may include: transferable work experience, skill sets, and qualifications. Total compensation also includes a comprehensive benefit package, including: medical, dental, vision, 401(k) plan, unlimited paid time off, generous parental leave plan, and others for mental and wellness support.

While we are a remote-first company, we have opened offices in New York City and the San Francisco Bay Area, as an option for those in those cities who wish to work in-person. For all other employees, there is a WFH monthly stipend to pay for co-working spaces.

More About Arize

Arize’s mission is to make the world’s AI work—and work for people. Our founders came together through a shared frustration: while investments in AI are growing rapidly across every industry, organizations face a critical challenge—understanding whether AI is performing and how to improve it at scale.

Learn more about what we're doing here:

https://techcrunch.com/2025/02/20/arize-ai-hopes-it-has-first-mover-advantage-in-ai-observability/

https://arize.com/blog/arize-ai-raises-70m-series-c-to-build-the-gold-standard-for-ai-evaluation-observability/

Diversity & Inclusion @ Arize

Our company's mission is to make AI work and make AI work for the people, we hope to make an impact in bias industry-wide and that's a big motivator for people who work here. We actively hope that individuals contribute to a good culture

• Regularly have chats with industry experts, researchers, and ethicists across the ecosystem to advance the use of responsible AI
• Culturally conscious events such as LGBTQ trivia during pride month
• We have an active Lady Arizers subgroup