AI Solution Architect - Agentic AI

We are seeking a highly experienced Architect to lead the design and development of multi-agent and agentic workflow systems. This role focuses on creating the orchestration layer, memory design, and tool-use permissions that ensure autonomous systems remain reliable and safe. The ideal candidate will have a strong background in software and AI architecture, particularly with multi-agent frameworks, and will be responsible for driving technical excellence across agentic system designs.

Key Responsibilities

- Design multi-agent architectures, including defining agent roles, coordination patterns, and handoff logic using frameworks such as LangGraph, AutoGen, and CrewAI.
- Own the design of agentic workflows encompassing task decomposition, tool invocation, memory and state management, and the integration of knowledge graphs for structured reasoning and long-term memory.
- Define and implement guardrails including tool-use permissions, kill switches, and escalation paths to manage agent failures effectively.
- Architect evaluation frameworks for agentic systems in collaboration with AI Assurance and Quality Assurance teams.
- Lead technical design reviews for agent-based solutions prior to development to ensure robustness and alignment with best practices.
- Design fallback mechanisms and human-in-the-loop processes for critical agent decision points.
- Mentor GenAI Engineers on architectural patterns and review their design work to maintain high standards.
- Present risk and benefit tradeoffs of agentic systems to both technical and business stakeholders, facilitating informed decision-making.
- Own architecture governance for agentic builds across the organization, ensuring consistency and compliance.
- Facilitate cross-team design reviews involving Security, Governance, and Platform Engineering teams to address interdisciplinary concerns.

Required Qualifications

- 10 to 15+ years of experience in software or AI architecture, with at least 2 years specifically focused on agentic or multi-agent systems.
- Deep expertise in agent orchestration frameworks and a thorough understanding of their failure modes, such as loops, deadlocks, and drift.
- Proven experience evaluating and architecting on managed agentic platforms, including Microsoft Azure AI Foundry Agent Service, AWS Bedrock Agents/Strands Agents, Google Vertex AI Agent Builder, as well as open-source orchestration tools like LangGraph, AutoGen, and CrewAI.
- Strong foundation in systems design principles, including distributed systems, state machines, and event-driven architectures.
- Experience designing memory architectures for long-running agents, incorporating vector and graph-based approaches.
- Familiarity with agent evaluation methodologies and benchmarking techniques.
- Working knowledge of guardrail and governance tooling specific to agentic systems.
- Ability to confidently defend architectural decisions under technical scrutiny from client architects.
- Skilled in writing clear Architecture Decision Records (ADRs) and documentation accessible to both engineers and executives.
- Demonstrated experience mentoring and leading technical teams beyond individual contributor roles.
- Comfortable being the accountable technical owner for agentic systems in production, including managing failures and incidents.

This role offers the opportunity to shape the future of autonomous agent systems by architecting solutions that balance innovation with safety and reliability. Candidates who thrive in collaborative, high-stakes environments and possess a passion for advancing multi-agent AI technologies will find this position rewarding.