← five9 / Principal Product Manager, Google Gemini Enterprise for Customer Experience (GECX)
brief / art_0nb5rUeQDBc
role
model
anthropic/claude-sonnet-4.6
created
2026-06-02T23:39
Company snapshot
Five9 is a publicly traded cloud contact center software provider (CCaaS) serving enterprise and mid-market customers globally, competing primarily with Genesys, NICE CXone, and Talkdesk. The company has made AI a central pillar of its strategy, building out intelligent virtual agents, agent-assist, and workflow automation capabilities. Five9 has an established strategic partnership with Google Cloud, and the GECX (Google Gemini Enterprise for Customer Experience) integration represents a high-priority joint go-to-market initiative. Based on the JD, Five9 operates distributed engineering and product teams across the US, Portugal, and India. Specific recent financial results or named internal initiatives beyond what the JD describes are not confirmed here.
Team stack
Based on the JD and Five9's public product surface: core platform is cloud-native SaaS (likely AWS and/or GCP given the Google partnership); AI layer integrates Google Gemini/GECX APIs including Agent Assist, Conversational Insights, and Quality AI; agentic orchestration likely uses ReAct/Chain-of-Thought patterns and multi-step tool-use frameworks (based on JD language); data/reporting stack likely includes BigQuery or similar (given GCP alignment); contact center integrations span voice, digital channels (chat, email, SMS), and CRM connectors; LLM application layer likely involves RAG architectures and prompt engineering (based on JD preferred qualifications); frontend tooling for agent desktop is likely React-based (inferred from industry norms, uncertain).
Likely questions (10)
| area | question | why |
|---|---|---|
| system_design | Walk us through how you would architect the integration between Five9's contact center platform and Google GECX Agent Assist — covering data flow, latency constraints, fallback handling, and how you'd ensure the agent desktop experience remains seamless. | The JD explicitly calls out designing and overseeing integration with Google's AI agents for multi-step autonomous actions; this tests whether the candidate can think end-to-end about a real-time, latency-sensitive AI integration. |
| domain | How would you define and measure the success of an AI-powered Agent Assist feature in a contact center environment? What metrics matter most to enterprise buyers, and how do you balance agent adoption with automation depth? | The JD requires deep domain expertise in contact center quality frameworks and CX tech stacks; this probes whether the candidate understands CCaaS-specific KPIs (AHT, CSAT, FCR, containment rate) beyond generic product metrics. |
| system_design | Describe how you would design a multi-agent orchestration system where Google Gemini agents autonomously handle complex, multi-step customer interactions — including how you'd handle agent handoff, context preservation, and failure recovery. | The JD specifically calls out 'multi-step actions autonomously across every consumer touchpoint' and lists agentic design patterns (ReAct, Chain-of-Thought, tool use, multi-agent orchestration) as preferred qualifications. |
| behavioral | Tell me about a time you led a complex strategic partnership between your company and a major technology partner. How did you align internal and external stakeholders, and what did you do when priorities diverged? | The JD emphasizes partnering closely with Google R&D and GTM teams on a strategic partnership — this is a core execution challenge for the role and tests influence-without-authority skills. |
| coding | You need to evaluate whether a new Gemini-powered Agent Assist feature is actually improving agent performance. Walk me through how you'd design an A/B test — including randomization unit, guardrail metrics, and how you'd handle the fact that agents may be aware they're being tested. | The JD requires 'track record of managing beta programs, running experiments (A/B testing)' — this tests statistical and experimental design rigor in a domain-specific context. |
| domain | How would you approach building an evaluation framework for an LLM-powered Conversational Insights product that analyzes customer call transcripts? What dimensions would you evaluate, and how would you handle hallucination or factual drift in quality scoring? | Conversational Insights and Quality AI are explicitly named GECX components; the JD lists 'responsible AI principles, evaluation frameworks' as something the candidate should champion — directly maps to candidate's aeval platform work. |
| behavioral | Describe a situation where you had to drive a 0-to-1 product from concept to GA in a matrixed organization with distributed teams. How did you maintain velocity and alignment across engineering, design, and GTM? | The JD calls out leading cross-functional teams across the US, Portugal, and India, and requires a 'proven track record of successfully launching and scaling Applied AI products from conception through growth.' |
| culture | Five9 is positioning itself as the undisputed leader in intelligent CX. How do you stay ahead of fast-moving competitors like Genesys and NICE in the AI agent space, and how do you decide when to build vs. partner vs. integrate? | The JD asks the candidate to 'conduct deep competitive and market analysis, identifying emerging trends in CX and autonomous agents' — this tests strategic market thinking and build/buy/partner judgment. |
| behavioral | Tell me about a time you championed responsible AI or safety principles in a product you shipped. How did you balance innovation speed with risk mitigation, and how did you bring skeptical stakeholders along? | The JD explicitly calls out 'responsible AI principles' as something the candidate should champion across the organization — this is a values and judgment signal Five9 is clearly screening for. |
| domain | Walk me through how you would build a business case and ROI model for a joint Five9-Google GECX deployment for a large enterprise contact center customer. What inputs would you need, what assumptions would you make transparent, and how would you present it to a CFO? | The JD explicitly requires candidates to 'build business cases and ROI models that demonstrate the joint value of Five9 and Google GECX' — this tests financial acumen and enterprise sales partnership skills. |
Talking points
- Platform infrastructure at scale with developer-facing impact: At Intuit, led the ICE platform to 675M+ engagements in FY23, scaled throughput from 6K to 50K TPS via rSocket migration supporting ~1.5M concurrent connections at sub-25ms TP99 — directly analogous to the latency-sensitive, high-concurrency demands of a real-time Agent Assist integration with Google GECX.
- Hands-on multi-agent orchestration and agentic design: Built OpenClaw, a production multi-agent orchestration framework with gateway protocol, subagent delegation, profile management, and session switching — directly maps to the JD's requirement for experience with 'AI agent orchestration frameworks' and 'multi-agent orchestration' agentic design patterns.
- AI evaluation rigor at the platform level: Built aeval, a local-first model evaluation platform covering factuality, reasoning, safety, and code generation with bootstrap confidence intervals, Welch's t-test, Cohen's d effect size, and CI/CD regression detection — positions candidate to immediately lead the 'evaluation frameworks and responsible AI principles' mandate in the JD.
- RL post-training and LLM depth: Built a full RL post-training workbench benchmarking 12 algorithms (PPO, GRPO, DPO, SimPO, etc.) across TRL, VeRL, OpenRLHF, and NeMo RL with live SSE metric streaming — demonstrates the LLM application depth (beyond surface-level prompt engineering) that Five9 needs to credibly partner with Google's Gemini R&D teams.
- 0-to-1 developer platform leadership with measurable business outcomes: At Intuit, delivered the ICE Self-Service platform reducing developer onboarding from 2–3 weeks to minutes, mitigating $1M+ in projected opex growth; implemented ICE Presence generating $480K/month in additional invoicing — demonstrates the ability to translate platform investments into quantified business cases, directly matching the JD's ROI modeling and GTM requirements.