Senior Applied AI Engineer

London United Kingdom Ref: 2038

Applied AI Engineer ×4 roles | Netherlands / UK | Hybrid

ABOUT THE COMPANY

Oxygen Digital is supporting a fast-growing B2B SaaS business in the industrial and operations space to hire four Applied AI Engineers across two product squads. The company is building genuinely AI-native features into a product used by large enterprise customers — maintenance teams, operations managers, and field technicians who rely on the platform day to day.

This is a build-and-ship mandate. The team is close to customers, moving fast, and looking for engineers who can raise the capability of everyone around them.

ROLE OVERVIEW

You will own a vertical slice of an AI-native product feature end-to-end — from interface to orchestration to deployment to observability. You will work in a mixed squad alongside existing product engineers and PMs, reporting into the Chief AI Officer.

Four engineers will be hired across two product squads. Occasional customer-site visits are part of the role, including industrial environments.

KEY RESPONSIBILITIES

• Design, build, and ship production LLM-powered product features — real users, real load, real failure modes
• Build and operate RAG pipelines that survive long-tail customer data, not just clean demos
• Design and ship agentic systems with tool calling, multi-step reasoning, agent memory, and human-in-the-loop guardrails
• Own the full stack: frontend (React/TypeScript), backend (Python/TypeScript), and deployment — commit to production
• Build and maintain evaluation harnesses for non-deterministic AI systems
• Instrument AI systems with proper observability and tracing
• Contribute to a mixed squad — bring non-AI-native colleagues up, engage directly with customers

PERSON SPECIFICATION

Essential — Full-Stack Engineering

• Genuinely full-stack: comfortable shipping at every layer — frontend, backend, and deployment infrastructure
• DevOps literate: can deploy your own work, reason about CI/CD, debug containers, read infrastructure-as-code. Fluent on Azure
• Strong Python in production codebases — you know the difference between a working demo and a reliable service

Essential — Applied GenAI

• Hands-on production GenAI experience post-2023: shipped LLM-powered features to real users at real load — not hackathons or internal tools
• RAG expertise: chunking strategies, embedding model selection, hybrid search, re-ranking, citation and provenance tracking
• Familiarity with Graph RAG and when it outperforms vector-only retrieval
• Agentic systems experience: tool calling, function orchestration, multi-step reasoning, retry and validation loops — frameworks matter less than the underlying judgement
• Eval-literate: able to design and maintain evaluation harnesses for non-deterministic systems — LLM-as-judge calibration, regression suites, agent trajectory scoring, cost-per-task tracking
• Prompt engineering at system level: structured outputs, JSON-mode reliability, token and context budget management
• Multi-provider fluency: shipped against Anthropic, OpenAI, and at least one open-weights model

Essential — Production Discipline

• Experience handling hallucinations, latency tails, rate limits, cost overruns, and silent failures in production AI systems
• AI observability and tracing: Langfuse, LangSmith, Braintrust, OpenTelemetry, or equaivalent
• Vector databases: Pinecone, Weaviate, pgvector, Qdrant — including embedding lifecycle management

Strongly Preferred

• Experience in a forward-deployed or customer-facing AI engineering role
• Day-to-day use of AI coding agents (Claude Code, Cursor, Windsurf) and a view on Spec-Driven Development
• Background in industrial, operations, maintenance, or vertical SaaS

Candidate Profile

• High agency — ships, doesn’t wait for perfect conditions
• Opinionated but coachable; cares whether the feature works, not just whether the code is elegant
• Comfortable in a mixed squad and willing to bring non-AI-native colleagues along
• Customer-curious — willing to join calls, visit sites, watch real users work
• Comfortable with ambiguity and evolving scope

HOW TO APPLY

Send your CV and a brief covering note to [email protected] quoting Applied AI Engineer in the subject line.

Salary: £110k – £120k + benefits
Location: London (fully remote) with some travel (20%)