Want to know if this job is worth applying to?
Argentina
Remote only
At ASAPP, our mission is simple: deliver the best AI-powered customer experience—faster than anyone else. To achieve that, we’re guided by principles that shape how we think, build, and execute. We value customer obsession, purposeful speed, ownership, and a relentless focus on outcomes. We work in tight, skilled teams, prioritize clarity over complexity, and continuously evolve through curiosity, data, and craftsmanship. We’re seeking technologists and problem solvers who thrive in fast-paced environments, love collaborating with great talent, and approach every day like it’s Day 1.
We're a globally diverse team with hubs in New York City, Mountain View, Latin America, and India. If you're driven by continuous learning, rapid pivots, and the challenges of building in a high-growth startup, we’d love to talk. This is more than a job—it’s a journey.
This is not a reporting role. You will own the tools that let builders and reviewers see, test, and trust how GenerativeAgent (GA) actually behaves — before it ever talks to a customer, and after. If builders can't confidently preview a configuration change, and reviewers can't efficiently audit what GA said, thought, and did in production, quality erodes silently and trust with our customers erodes with it. Making agent behavior visible, testable, and reviewable at scale is your product.
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.
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.
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.
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
