Staff Product Manager (Runtime Team)

present of AI; Greater London; Hybrid; GBP 110,000 - 150,000; Full time

Job summary

PolyAI is hiring a Senior Product professional to own the platform layer that underpins our voice agents in production. This role focuses on how conversations are served, how models are hosted and served, and how data moves and is stored, driving direction and tradeoffs rather than managing a feature backlog.

A strong candidate brings years of product work on infrastructure, ML or data platforms, a solid grasp of latency budgets and serving economics, and the ability to advocate for backend

Qualifications

• Years of product work on infrastructure, ML or data platforms.
• Grasp of latency budgets and serving economics.
• Ability to argue for backend investment with no visible customer surface.

Responsibilities

• Owns the platform layer beneath voice agents, including serving and hosting models.
• Decide data flow and storage for conversations and models.
• Set direction and make tradeoffs, not manage feature backlog.

Skills

Infrastructure product ownership

Inference and serving tradeoffs

Latency and cost metrics

Data platform residency

Cross functional influence

Job description

Product · London, United Kingdom · Senior level

This role owns the platform layer beneath the company's voice agents: how conversations are served in production, how models are hosted and served, and how conversation and model data moves and is stored. A strong candidate brings years of product work on infrastructure, ML or data platforms, a working grasp of latency budgets and serving economics, and the ability to argue for backend investment that has no visible customer surface. The job is about setting direction and making tradeoffs, not managing a feature backlog.

How it fits the AI era

The product is AI itself, so the work sits close to how models are served and how agents perform in live conversations. Much of the judgment stays human: weighing cost against latency and reliability, deciding what to build versus buy, and setting how performance is measured in production. AI tools may speed up analysis and drafting, but the calls on platform risk and data handling remain with people.

How it builds AI experience

Working here means gaining hands on exposure to inference and model serving at scale, including self hosted and third party paths, plus evaluation of latency, throughput and cost per conversation. A candidate can build this by studying how serving infrastructure works and by showing past decisions on data pipelines, residency and access control. Familiarity with how dialog agents run in production would help, though the posting centers on infrastructure judgment rather than model building.

Worth showing

• Infrastructure product ownership
• Inference and serving tradeoffs
• Latency and cost metrics
• Data platform and residency
• Cross functional influence

This is our own read, written from the employer's posting and the company's Spotlight. The employer's posting is the source of truth for the role.

About PolyAI

PolyAI builds dialog agents that handle critical enterprise customer conversations and resolve them without escalation, per the company.