LLM Agentic AI Lead

ROLE MANDATE

A senior, hands-on lead accountable for implementation quality, agent evaluation, and secure production readiness. Owns implementation decisions within agreed architecture and risk policies. Partners with Architecture on design exceptions, Security on controls, and Product on priorities and acceptance criteria. Translates architecture into working solutions and develops engineers through coaching and code review.

WHAT SUCCESS LOOKS LIKE

Reliable agents meet agreed task-success, security, latency, and cost targets, with predictable delivery, faster recovery, and increasing pod independence.

CORE RESPONSIBILITIES

• Lead technical delivery, resolve trade-offs, and manage dependencies.
• Build agent orchestration, tool integrations, and context/state handling.
• Own agent evaluations, regression tests, and release acceptance criteria.
• Implement security controls, guardrails, and human approval paths.
• Own tracing, recovery, rollback, and cost/latency with platform partners.
• Set code standards and coach engineers through hands-on delivery.

SUCCESS PROFILE

• Experience shipping and operating production agentic AI solutions.
• Strong software engineering, tool calling, and context management.
• Practical depth in evaluations, observability, and secure execution.
• Sound judgment on agent vs. deterministic workflow trade-offs.
• Hands-on leadership, stakeholder communication, and coaching.

Security and guardrail accountability:

Enforce least-privilege tool access, protect sensitive data, address prompt injection, validate tool inputs and outputs, and define human approval and escalation paths. Agree risk policies with Security and implement controls within the approved architecture.