Accenture

Agentic Orchestration Engineer

Accenture
ConsultingSingaporeHybridPosted 2 weeks ago

About the role

AI summarised

Design and build production-grade multi-agent systems that power AI-driven workflows at scale. You will architect and implement advanced orchestration patterns, lifecycle management, and observability for agents, enabling reliable, secure, and efficient AI solutions across complex enterprise environments.

ConsultingHybridOther Functions

Key Responsibilities

  • Design and implement production-grade multi-agent orchestration frameworks based on a multi-tier pattern (e.g., Tier 1 routers, Tier 2 domain specialists, Tier 3 utility agents).
  • Build and maintain complex workflow orchestration using state machines (e.g., LangGraph) to manage agent coordination, context, and control flow.
  • Define and manage the full agent lifecycle, including registration, configuration, versioning, deprecation, and runtime governance.
  • Implement robust observability for agents and workflows, leveraging tools such as LangSmith and Datadog for tracing, logging, metrics, and performance insights.
  • Develop scalable, asynchronous Python services that integrate with LangChain/LangGraph and other orchestration components.
  • Collaborate with AI engineers, solution architects, and product teams to translate business workflows into orchestrated multi-agent solutions.
  • Ensure reliability, security, and resilience of agentic systems in production, including error handling, fallback strategies, and monitoring.
  • Contribute to and advocate for best practices, patterns, and reusable components for multi-agent orchestration across the organization.

Requirements

  • Strong hands-on experience building AI or automation systems using Python.
  • Practical experience with LangChain, LangGraph, or similar frameworks for agent or workflow orchestration.
  • Solid understanding of multi-agent system design, orchestration patterns, and state machines.
  • Experience with asynchronous programming and event-driven architectures in Python.
  • Familiarity with observability practices and tools (e.g., LangSmith, Datadog, or equivalent) for monitoring complex workflows.
  • Demonstrated ability to design and implement production-ready services with a focus on robustness, scalability, and maintainability.
  • Strong collaboration and communication skills, with the ability to work closely with cross-functional technical and product teams.