Production-Grade AI Agent Architect

Accenture PLC; Greater London; Presencial; GBP 90.000 - 130.000; Jornada completa

Descripción de la vacante

Accenture PLC seeks an AI Engineer (Agentic/Applied) to design, build, and deploy production-grade agentic AI systems across the enterprise tech stack. You will collaborate directly with client engineering teams, lead technical design sessions, and develop reusable patterns that scale beyond individual engagements.

This role emphasizes hands-on production experience, governance of eval harnesses, and a strong focus on reliability, observability, and cost control across multiple concurrent

Formación

• Production-grade software engineering experience in live environments.
• Experience deploying agentic AI solutions in production settings is required.
• Hands-on with agentic orchestration frameworks and LLM APIs in production.

Responsabilidades

• Architect agentic systems at enterprise scale with multi-agent orchestration and RAG pipelines.
• Define standards for RAG pipelines, embeddings, and quality benchmarks.
• Set multi-LLM integration standards across providers like OpenAI, Anthropic, Vertex AI.
• Own LLMOps: eval strategy, prompt governance, observability, and cost controls.
• Lead client engineering engagements, run architecture sessions, and deliver PoCs.
• Publish reusable patterns and accelerators to scale across engagements.
• Define program-level AI impact metrics and present to senior stakeholders.

Conocimientos

Python

Java

LLMOps

LangGraph

CrewAI

AutoGen

LLM APIs

Production systems

Distributed systems

Observability

Herramientas

Kubernetes

Docker

Terraform

Helm

Descripción del empleo

Accenture PLC seeks an AI Engineer (Agentic/Applied) to design, build, and deploy production-grade agentic AI systems across the enterprise tech stack. You will collaborate directly with client engineering teams, lead technical design sessions, and develop reusable patterns that scale beyond individual engagements.

This role emphasizes hands-on production experience, governance of eval harnesses, and a strong focus on reliability, observability, and cost control across multiple concurrent