Senior Product Manager, Observability Tools

OpenKrill; United States; Remote; USD 125,000 - 180,000; Full time

Benefits offered by this job

Competitive pay

Stock options

Family healthcare

Mac computer

Flexible benefits

3 weeks vacation

English lessons

Job summary

ASAPP is seeking a senior Product Manager to own end‑to‑end product areas around simulation, evaluation, and review tooling for AI agents. You will drive scenario generation, evaluation surfaces, and the preview/inspection workflows used by builders and customers to ensure quality before GA release.

You will collaborate with research and engineering to ground product decisions in technical reality, measure outcomes, and communicate clearly with cross‑functional teams and external customers.

Qualifications

• 5+ years of product management experience in enterprise B2B SaaS with testing/quality focus.
• Direct experience building products around LLMs, AI agents, or applied AI systems and evaluating model outputs.
• Experience shipping tools that simulate, test, or preview system behavior before production.
• Experience building review or audit tooling for complex, unstructured content.

Responsibilities

• Own scenario creation and lead product for tools that generate/curate test scenarios.
• Own evals surfaces, including mistake monitoring and structured data evals at scale.
• Own the Previewer tool used to see how configurations appear in simulated conversations.
• Lead the Conversation Explorer surface reviewers who audit full production conversations.

Skills

Product management

Testing & quality

AI/LLM product experience

Stakeholder collaboration

Experimentation

Tools

Evaluation frameworks

Job description

What You Will Do

• Own scenario creation via the simulation agent. Lead product for the tools that let builders generate and curate realistic test scenarios, so configuration changes can be validated against representative conversations before they ship.
• Own our evals surfaces, including mistake monitoring and structured data evals. Define how we detect, surface, and categorize GA mistakes at scale, and how structured data extraction is evaluated for accuracy — turning noisy conversational output into clear, actionable quality signals for builders and customers alike.
• Own the Previewer. Own the tool builders rely on to see exactly how their configurations show up in real simulated conversations, closing the loop between making a change and understanding its downstream impact.
• Own Conversation Explorer. Lead the primary surface reviewers use to read through full production conversations — including agent thoughts and actions — balancing depth for investigative review with speed for high-volume auditing.
• Build the case for simulation-driven quality. Partner with Sales, Customer Success, and Research to show customers and internal stakeholders how pre-production testing and post-production review compound into fewer mistakes, faster iteration, and more defensible GA performance.
• Partner deeply with Research and Engineering on the underlying agent architecture — reasoning traces, tool calls, scenario generation methods — so product decisions are grounded in what’s technically sound and what actually predicts real‑world behavior.
• Manage the trust layer of the review experience. Proactively identify and resolve issues that erode confidence in what builders and reviewers see — desyncs between simulated and real behavior, incomplete traces, unclear mistake categorization — and design the UX and guardrails that keep them confident in what they’re looking at.

Who You Are

We evaluate candidates against a set of principles. These aren’t interview talking points — they describe how PMs are expected to operate here every day.

• Growth & Abundance Mindset: You approach new domains with curiosity and learn fast. You see collaboration as additive, not competitive. When a bet doesn’t work, you extract the signal and move on.
• Extreme Ownership: You take responsibility to deliver. You don’t wait to be asked — you identify the gap and close it. You are the PM; the outcome is yours.
• High-Velocity Outcomes: You measure yourself by outcomes, not output. You distinguish between shipping a feature and moving a metric. You can answer "what changed for the customer?" for everything on your roadmap.
• Effective Communication, Collaboration, and Trust: You influence without authority across engineering, research, design, sales, and customer success. You overcommunicate to keep stakeholders aligned.
• Real-Time Feedback: You give and receive feedback continuously rather than saving it for quarterly reviews. When something is wrong, you say so directly and constructively.

What You Will Need

• 5+ years of product management experience, with meaningful time spent on testing, quality, or review/analytics products in enterprise B2B SaaS.
• Direct experience building products around LLMs, AI agents, or other applied AI systems — you can reason about agentic behavior, prompt and scenario design, or model output evaluation well enough to make sharp product tradeoffs, not just talk about AI at a high level.
• Experience shipping tools that simulate, test, or preview system behavior before it reaches production, including the judgment to know when simulated behavior is a reliable proxy for the real thing and when it isn’t.
• Experience building review or audit tooling for complex, unstructured content, and turning ambiguous quality signals into clear, defensible categorizations.
• Comfort partnering with research and engineering teams on system internals — reasoning traces, tool‑use logs, evaluation methodology — to ground product decisions in technical reality.
• Track record of running discovery with internal or external users and translating it into structured requirements — evals, quality metrics, or review workflows.
• Excellent written and verbal communication — you can write a clear PRD, a release announcement, and comfortable presenting directly to customers.

What We Would Like to See

• Experience in contact centers, conversational AI, or customer experience (CX) analytic - a plus.
• Experience with evaluation frameworks or methodologies for generative AI systems (offline evals, human‑in‑the‑loop review, automated mistake detection).
• Prior experience in a startup or fast‑scaling environment where the role’s scope evolved as the company did

Benefits

• Competitive compensation
• Stock options
• Healthcare for the family group
• Mac equipment
• USD 150 per month in flexible benefits
• 3 weeks of vacation
• English lessons

ASAPP is committed to creating a diverse environment and is proud to be an equal opportunity employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, gender, gender identity or expression, sexual orientation, national origin, disability, age, or veteran status. If you have a disability and need assistance with our employment application process, please email us at [e-mail] to obtain assistance. #LI-SL1 #LI-Remote