Requirements
Requires 10-15+ years of software/AI architecture experience, with at least 2 years specifically in agentic/multi-agent systems. Deep knowledge of agent orchestration frameworks, their failure modes, and managed agentic platforms (Azure AI Foundry, AWS Bedrock Agents, Google Vertex AI Agent Builder, open-source options like LangGraph, AutoGen, CrewAI) is essential. Strong systems design fundamentals, experience with memory architectures, agent evaluation, and guardrail tooling are also required. Experience mentoring technical teams and leading architecture governance is necessary.
Description
ABOUT:Architects multi-agent and agentic workflow systems — the orchestration layer, memory design, and tool-use permissioning that keep autonomous systems reliable and safe.KEY RESPONSIBILITIESDesign multi-agent architectures — agent roles, coordination patterns, handoff logic (LangGraph, AutoGen, CrewAI)Own agentic workflow design: task decomposition, tool-calling, memory/state management, including knowledge graphs for structured reasoning and long-term memoryDefine guardrails: tool-use permissioning, kill switches, escalation paths for agent failureArchitect the evaluation framework for agentic systems in partnership with AI Assurance/QALead technical design reviews for agent-based solutions before they reach buildDesign fallback and human-in-the-loop patterns for high-stakes agent decisionsMentor GenAI Engineers on architecture patterns and review their designsPresent agentic system risk/benefit tradeoffs to client technical and business stakeholdersOwn architecture governance for agentic builds across the practiceFacilitate cross-team design reviews involving Security, Governance, and Platform EngineeringREQUIREMENTS & SKILLS10–15+ yrs software/AI architecture, with 2+ yrs specifically in agentic/multi-agent systemsDeep knowledge of agent orchestration frameworks and their failure modes (loops, deadlocks, drift)Evaluates and architects on managed agentic platforms as per area of specialization including Microsoft Azure AI Foundry Agent Service, AWS Bedrock Agents/Strands Agents, and Google Vertex AI Agent Builder; open-source orchestration (LangGraph, AutoGen, CrewAI) etc.Strong systems-design fundamentals — distributed systems, state machines, event-driven architectureExperience designing memory architectures (vector + graph-based) for long-running agentsFamiliarity with agent evaluation methodologies and benchmarking approachesWorking knowledge of guardrail/governance tooling for agentic systemsCan defend an architecture decision under adversarial technical challenge from a client's own architectsWrites clear ADRs and architecture documentation that both engineers and executives can followExperience mentoring/leading a technical team, not just designing soloComfortable being the accountable technical owner when an agentic system fails in production