Principal Applied AI Engineer

Zebra People | B Corp™; Greater London; On-site; GBP 120,000 - 180,000; Full time

Job summary

Zebra People partners with a large UK organisation investing in Applied AI to build a senior engineering capability focused on productionising AI systems. They seek a Principal Applied AI Engineer who blends software engineering with hands‑on experience building agentic AI systems.

This is a technical leadership role, but very hands‑on, owning complex AI projects while shaping reusable tooling, design patterns and engineering standards across the wider organisation.

Qualifications

• Strong Python/software engineering experience.
• Experience shipping LLM or agentic systems into production.
• Deep understanding of agent architectures and orchestration.
• Strong experience with AI evaluation / evals.
• Context engineering, RAG, retrieval or memory experience.
• Strong system design and architecture capability.
• Experience leading technically complex projects.

Responsibilities

• Designing and delivering production‑grade AI agents and multi‑agent systems.
• Owning complex AI projects from architecture through to production.
• Building agent tooling, orchestration, memory/context and retrieval systems.
• Working with MCP / Model Context Protocol.
• Designing evaluation frameworks, golden datasets and quality metrics.
• Building reusable AI components and engineering patterns.
• Helping shape approaches to spec‑driven development and AI coding agents.
• Providing technical leadership, mentoring and architecture guidance.

Skills

Python

LLM/Agentic systems

System design

Technical leadership

AI evals

RAG retrieval memory

Agent architectures

Project leadership

Tools

LangGraph

LangChain

LlamaIndex

AWS

Claude Code

Codex

GitHub Copilot

OpenCode

MCP

Job description

I'm working with a large UK organisation that is making a significant investment in Applied AI and building out a senior engineering capability focused on taking AI systems from experimentation into production.

They're looking for a Principal Applied AI Engineer who combines strong software engineering fundamentals with hands‑on experience building and operating agentic AI systems.

This is a technical leadership role, but very much a hands‑on one. You'll spend the majority of your time owning complex AI projects, while also helping establish reusable tooling, design patterns and engineering standards across the wider organisation.

What you'll be doing

• Designing and delivering production‑grade AI agents and multi‑agent systems
• Owning complex AI projects from architecture through to production
• Building agent tooling, orchestration, memory/context and retrieval systems
• Working with MCP / Model Context Protocol
• Designing evaluation frameworks, golden datasets and quality metrics
• Building reusable AI components and engineering patterns
• Helping shape approaches to spec‑driven development and AI coding agents
• Providing technical leadership, mentoring and architecture guidance

What they're looking for

The biggest requirement is a strong engineering background.

You’ll ideally have:

• Strong Python/software engineering experience
• Experience shipping LLM or agentic systems into production
• Deep understanding of agent architectures and orchestration
• Strong experience with AI evaluation / evals
• Experience with MCP
• Context engineering, RAG, retrieval or memory experience
• Strong system design and architecture capability
• Experience leading technically complex projects

They're flexible on specific frameworks. LangGraph, LangChain, LlamaIndex and similar are all relevant, but they're particularly interested in engineers who understand what's happening underneath the framework rather than relying entirely on abstractions.

Experience with AWS, regulated environments, model fine‑tuning or voice AI is useful, but not essential and other tools will be fine.

Experience using coding agents such as Claude Code, Codex, GitHub Copilot or OpenCode is also valuable.

The Applied AI capability is still being built, so this isn’t a case of joining a mature function and maintaining what already exists.

You'll have genuine influence over how AI systems are engineered, evaluated and deployed across a large organisation, while working with significant data, real‑world production use cases and the resources to build things properly.