Agentic AI Architect

Responsibilities:

• Lead the end-to-end design and architecture of autonomous and semi-autonomous AI agents (including multi-agent systems) that orchestrate complex, cross-functional business workflows.
• Own strategic and hands-on technical decisions for models, orchestration frameworks, vector databases, runtime platforms, observability tooling, and policy/guardrail layers to balance performance, safety, and cost.
• Drive the full lifecycle of agentic AI solutions, from PoCs through pilots, production rollout, and continuous optimization.
• Build, mentor, and retain a high-performing team of ML/AI engineers, agent/prompt engineers, and product partners focused on delivering high-value agentic use cases.
• Define, implement, and enforce governance for safety, security, privacy, and ethics of autonomous agents, including human-in-the-loop controls, escalation paths, and clear accountability for decisions.
• Partner with security, risk, legal, and compliance functions to embed organizational policies and external regulations into agent behavior, access controls, and data handling.
• Design and implement robust monitoring and observability for agents, covering reliability, accuracy, latency, drift, hallucination patterns, business KPIs, and critical failure modes, with clear remediation runbooks.