Senior AI Engineer in London

Architect and lead the implementation of multi-agent systems using Google AI SDKs (Vertex AI Agent Builder), LangGraph, CrewAI, and other emerging orchestration frameworks. Design and build stateful, tool-augmented agents capable of advanced reasoning, long-term planning, and autonomous execution. Develop and document agent orchestration patterns including planner-executor, supervisor-worker, and hierarchical agent structures. Implement sophisticated memory systems (short-term, long-term, and cross-session contextual memory). Enable seamless cross-agent communication and multi-modal coordination. Lead the delivery of production-grade LLM applications: RAG pipelines, specialised agents, and developer copilots. Integrate diverse tools, enterprise APIs, and legacy systems into agentic workflows. Design robust system prompts, dynamic routing logic, and AI guardrails using Vertex AI Model Garden or Azure AI Studio. Drive optimisation of AI workflows for latency, token cost, and output quality. Develop and own reusable AI microservices, agent frameworks, and standardised APIs. Contribute to core AI platform capabilities including model routing, centralised observability, and safety filters. Define and enforce engineering standards and best practices for AI development across the team. Deploy and manage agent-based systems on GCP, Azure, and/or AWS using Docker, Kubernetes (GKE/AKS/EKS), and Cloud Run. Implement comprehensive monitoring and observability using Vertex AI Inspector, LangSmith, or Azure Monitor. Drive incident response and post-mortems for production AI system failures. Act as a technical lead on key AI engineering workstreams, shaping architecture and approach. Mentor and support more junior AI engineers through code review, design discussions, and pair programming. Collaborate with Principal AI Engineer and cross-functional teams (data, product, delivery) to align AI engineering with business outcomes. Stay at the forefront of the rapidly evolving agentic AI landscape and bring new approaches into the team.

Requirements

• 5–8 years of software engineering experience with at least 3 years focused on LLM-based or AI systems in production.
• Proven track record building and shipping RAG pipelines, autonomous agents, and multi-step reasoning chains.
• Strong hands-on experience with Google AI SDKs, Vertex AI, and/or Azure AI services.
• Deep proficiency in orchestration stacks: LangGraph, CrewAI, LlamaIndex, Haystack, or comparable frameworks.
• Expert-level Python; strong backend development skills (FastAPI, Go, or Node.js).
• Deep understanding of agent design patterns: planning, reflection, memory, and tool-use.
• Experience integrating complex enterprise APIs and event-driven systems into agentic workflows.
• Proven ability to trace, debug, and improve non-deterministic, multi-step AI reasoning pipelines.
• Strong instinct for building resilient, observable, and production-ready AI systems.
• Strong familiarity with GCP and/or Azure core services: GKE, Cloud Run, Azure AI services.
• Infrastructure as Code: Terraform or Pulumi.
• CI/CD: experience building automated evaluation and deployment pipelines for AI models.

ATS Optimization Keywords

• Hard Skills: Python, FastAPI, Go, Node.js, Google AI SDKs, Vertex AI, Azure AI services, LangGraph, CrewAI, RAG pipelines.
• Soft Skills: leadership, mentoring, collaboration, problem-solving, communication, design discussions, code review, incident response, optimisation, best practices.