Staff Product Manager (Evals)

We're looking for a Staff Product Manager to own evaluations for AI agents at Workato — both the internal framework that helps our teams ship better AI features, and the customer-facing tools that let builders assess and improve the agents they create. This is a role with a dual mandate. Internally, you'll establish how Workato evaluates agent quality, starting with Agent Studio and expanding to other teams shipping AI capabilities. Externally, you'll build the evaluation experience that helps business technologists understand why their agents succeed or fail — and what to do about it. The right person for this role has actually written evals. You've built test suites, designed evaluation criteria, and debugged agent failures in the trenches. You know the gap between "eval theory" and "eval reality," and you can translate that practitioner knowledge into products that work for both technical teams and non-technical builders.

In this role, you will also be responsible to:

• Define and own the evaluation framework for Workato's internal AI agent features, driving adoption across teams starting with Agent Studio

• Build the customer-facing evaluation experience — how builders test, measure, and improve agents they create on Workato

• Make hard calls about what evaluation complexity to expose versus abstract, balancing rigor with approachability

• Partner closely with the Build Experience PM to ensure evaluation is integrated into the builder journey, not bolted on

• Work with ML engineers and platform teams to ground the framework in technical reality while keeping it accessible

• Establish metrics for what "good" looks like — both for internal agent quality and for customer evaluation adoption

• Spend significant time with customers understanding where they struggle to assess agent performance and what mental models they bring

• 7+ years in Product Management

• Hands-on experience writing evaluations for AI/ML systems (agents, LLMs, or similar)

• Track record of shipping technical products to both internal and external users

• Experience driving adoption of frameworks or practices across engineering teams

• Strong written and verbal communication skills

• Bachelor's degree or equivalent experience

• Practitioner depth in evaluations. You've written evals yourself — built test suites, designed rubrics, debugged why agents underperformed. You understand evaluation methodology not only from reading about it, but from doing it. You have opinions about what works, what doesn't, and where current approaches fall short.

• Strong product management experience. You've shipped products, driven roadmaps, and led cross-functional teams. You know how to translate technical capabilities into user value and write specs that don't leave details to chance.

• Technical translation ability. You can take complex evaluation concepts and make them accessible to business technologists without dumbing them down. You understand the difference between hiding complexity and organizing it.

• Internal influence skills. You've driven adoption of frameworks, practices, or tools across teams. You can be a credible partner to ML engineers while advocating for what internal teams actually need.

• Greenfield comfort. You've defined products from ambiguity — scoped v1s, made bets with incomplete information, and iterated based on what you learned. You don't need an existing playbook to be effective.

• B2B product sensibility. You see enterprise conventions as problems to solve, not constraints to accept. You're drawn to products that make complex workflows feel elegant.

• Experience with agent architectures, RAG systems, or LLM application development

• Background in ML engineering, solutions architecture, or technical program management before PM

• Experience building developer tools or platform products

• Familiarity with evaluation frameworks (e.g., human eval pipelines, automated benchmarks, red-teaming)
  

(REQ ID: 2538)