Staff Conversation Designer (Support)

Description

The company is now part of a larger global group following a merger with a sister brand, so you may work with colleagues and systems across both businesses.

The product team shapes how millions of people use the service, turning complicated marketplace problems into simple, high-impact solutions by combining data, design and engineering.

This London role sets design direction for AI-driven support across the group's brands. You bring AI skill, hands-on technical practice and an editorial view to design conversations powered by the company's LLMs, deciding what to say, what to leave out and how to keep each exchange clear, helpful and trustworthy.

Responsibilities

• Set support strategy across customer journeys by linking what people need now, what the business can promise and what operations can deliver. Help steer roadmaps with product, engineering and operations on where to invest, what comes first and how to weigh quality, speed and impact.
• Own chat, voice and proactive guidance from definition through launch, aiming to prevent or fix problems before customers ask for help. Write multi-turn specs covering context, interruptions, shifting intent, clarification, error recovery, fallbacks and human handoffs, without making users repeat themselves.
• Invent and test new conversational patterns, personas and support experiences using LLMs and code. Work with engineering to define the information, actions and human judgement each experience needs, and use prototypes and user feedback to decide what to pursue.
• Set the editorial direction for voice, tone and behaviour across brands, markets and channels. Write and refine system prompts and persona definitions that shape model behaviour, adapting to local voices, policies and access needs while keeping answers useful and promises accurate.
• Create evaluation datasets, testing plans, scoring rubrics and annotation guidelines. Calibrate judgements so the team assesses quality consistently, and combine results with research, performance data and reviews of hundreds of transcripts to improve accuracy, resolution quality and customer effort.
• Identify conversational risks such as hallucinations, harmful or biased replies and out-of-scope requests. Design, build and test guardrails covering sensitive data, limits on what the system can do, and when to admit uncertainty or bring in a person.
• Grow the wider studio's skills through mentoring, teaching and shared tools. Build reusable prompt guidance and evaluation tools with design and cross-functional peers, and drive their adoption across teams.

Requirements

• A portfolio of work showing editorial and conversational judgement, drawn from conversational systems, content programmes or shipped products. Show your personal contribution, the direction you set, the options you explored and the measured results. Different formats are welcome if they make your thinking, execution and influence clear.
• Experience setting direction across teams and shaping product priorities, including resolving ambiguity and balancing user needs, commercial goals and technical constraints.
• A strong editorial point of view backed by excellent writing and editing. You can explain why a conversation needs a given message, sequence or tone, cut what gets in the way and give specific feedback that improves others' work. Your background may be in editorial, content design, linguistics, product design or another field, as long as you apply that judgement to interactive experiences.
• Practical LLM and code skills, shown through working AI prototypes and direct investigation of model behaviour. You can tell conversational problems apart from limits in data, tools or the wider system.
• The ability to connect individual conversations to the wider service, including user intent, context, available actions and agent workflows. Your decisions account for accessibility, localisation, privacy and cases that need human intervention.
• Experience writing conversation evaluations and annotating model-training data, with evidence that qualitative and quantitative findings improved model behaviour and informed product decisions.
• A track record of raising other teams' design quality through mentoring, critique, and shared tools or practices they have adopted.

Conditions

The role is based in London.

Benefits

• Work on problems that matter in a three-sided marketplace that keeps changing.
• Tackle large-scale challenges on a marketplace that runs around the clock and reaches millions of people every day.
• See results quickly: ship, test and improve ideas in a low-hierarchy setting where you own the outcome.
• Take on big, unclear problems and advance your career with solid support.
• Work in a culture that values high standards, collaboration, flexible working and continuous learning.
• Pay is competitive, and equity is available, so you share in the impact you create.
• The tech blog offers a closer look at how the team builds its products.