Full-Stack AI Engineer: Build End-to-End AI in Production

commercetools; Greater London; Hybrid; GBP 120,000 - 180,000; Full time

Benefits offered by this job

Comprehensive health benefits

Learning and development

Family Leave Plus

Equity participation program

Job summary

commercetools is looking for a hands-on AI engineer who owns end-to-end AI capabilities—from data and context, through agent orchestration, to the generative UI a user sees and trusts. You’ve shipped and operated full-stack apps in production, so backend, frontend, data, and ML aren’t four teams you coordinate with; they’re four things you do.

You will build agent loops and orchestration, the integration layer (via MCP) for secure calls to backend systems, render model output into live UI, and

Qualifications

• 5+ years shipping production software.
• 2+ building with LLMs in production (real users).
• Hands-on experience with UI protocols (A2UI, AG-UI, MCP-UI).
• Experience with MCP, Python and TypeScript/Node + React.
• Fluent with retrieval/embeddings and SQL.

Responsibilities

• Build agent loops and orchestration: tool use, structured outputs, policy enforcement.
• Build the integration layer (via MCP) enabling secure calls to backend systems and business APIs.
• Render model output into live, streamed UI using generative/UI protocols.
• Ground models in real data: retrieval/embeddings and feedback loops.
• Treat evals as a first-class system and deliver production features from prompt-based ideas.

Skills

Full-stack development

LLMs in production

UI protocols

Python

TypeScript

React

MCP

Retrieval/embeddings

SQL

Observability

Tools

Python

TypeScript

React

MCP

Job description

commercetools is looking for a hands-on AI engineer who owns end-to-end AI capabilities—from data and context, through agent orchestration, to the generative UI a user sees and trusts. You’ve shipped and operated full-stack apps in production, so backend, frontend, data, and ML aren’t four teams you coordinate with; they’re four things you do.

You will build agent loops and orchestration, the integration layer (via MCP) for secure calls to backend systems, render model output into live UI, and