Production-Grade AI Agent Architect

Accenture UK & Ireland; Birmingham; On-site; GBP 90,000 - 140,000; Full time

Job summary

Accenture UK & Ireland is seeking an AI Engineer (Agentic/Applied) to design, build, and deploy production-grade agentic AI systems across the enterprise tech stack. You will work with client engineering teams, lead technical design sessions, and create reusable patterns that scale beyond engagements.

The role requires hands-on experience implementing agentic AI solutions in production, with strong Python/Java backend skills, and familiarity with LLM APIs from multiple providers.

Qualifications

• Production agentic AI systems experience is required.
• Experience with agentic orchestration frameworks (LangGraph, CrewAI, AutoGen).
• Direct production experience calling LLM APIs (OpenAI, Anthropic, Vertex AI).
• Cloud-native engineering with Kubernetes, Docker, microservices, serverless, CI/CD, IaC.

Responsibilities

• Architect and govern production-grade agentic systems at scale.
• Define RAG pipeline standards and embedding strategies.
• Set multi-LLM integration standards across providers.
• Own LLMOps and observability across programs.
• Lead client engineering engagements and drive architecture sessions.
• Publish reusable patterns and accelerators for scalability.
• Define programme-level AI metrics and communicate value to stakeholders.

Skills

Production software engineering

Agentic AI in production

LLM APIs in production

Kubernetes

Docker

CI/CD

IaC

Python

Java

Team leadership

Tools

LangGraph

CrewAI

AutoGen

Job description

Accenture UK & Ireland is seeking an AI Engineer (Agentic/Applied) to design, build, and deploy production-grade agentic AI systems across the enterprise tech stack. You will work with client engineering teams, lead technical design sessions, and create reusable patterns that scale beyond engagements.

The role requires hands-on experience implementing agentic AI solutions in production, with strong Python/Java backend skills, and familiarity with LLM APIs from multiple providers.