Senior Applied AI Engineer, Product (London)

Overview

In this hybrid role, you’ll own AI features end to end and serve as the bridge between backend systems and applied AI. You’ll build the surrounding infrastructure the AI relies on—ensuring reliability, latency, and real-time performance. You’ll evaluate and measure impact, iterate from real conversations, and collaborate across Product, Design, Support, and other engineering teams. This is a high-impact opportunity to shape AI-powered experiences used by thousands of salons and clients globally.

Responsibilities

• Own AI features end to end: design, build, evaluate, ship, and sustain for real users
• Build the system around the model: services, tooling, and integrations for real work (availability checks, bookings, payments, escalation)
• Prove it works: create evals, graders, and production checks and improve accuracy
• Engineer the context: determine model knowledge and tool access
• Work in real time: balance latency, cost, and quality
• Learn from real conversations: analyze transcripts and telemetry to drive changes
• Collaborate with Product, Design, Support and other engineers through the end-to-end lifecycle
• Influence architecture, AI practices, and engineering practices across the team
• Operate in a fast feedback loop with deploys behind feature flags
• Work across the stack; there is no strict AI/backend split

Key requirements

• 5+ years of experience building and running backend services in production
• Shipping products built on top of LLM APIs or agent frameworks with real user impact
• Experience writing evaluations for non-deterministic systems (test sets, graders, LLM-as-judge)
• Understanding of context and tool design (retrieval, tooling, MCP)
• Attention to correctness in distributed systems (idempotency, retries, failure handling)
• Experience with event-driven systems and cost/latency tradeoffs
• Product sense and customer-focused mindset
• Use of AI coding tools and ability to build own workflows; review of outputs before shipping
• Proficiency in at least one backend language; willingness to learn Elixir quickly

• Product sense
• Pragmatic approach
• Team collaboration and cross-functional communication
• LLMs and agent frameworks (LangGraph)
• Elixir and TypeScript services
• Postgres

…