Lenovo Digital Trust Lab is hiring a Staff Agentic AI Security Engineer to set the technical direction for our AI-powered security agents and lead their development across the team. As one of our most senior engineers, you will own the hardest, highest-leverage agent systems end-to-end, define the architecture and standards the rest of the team builds on, and act as a force multiplier — working directly with frontier models to solve security problems at a scale and level of autonomy that has not been done before.

Job Responsibilities:

• Set the technical vision and architecture for the agentic security platform — make foundational design decisions (planning, orchestration, tool use, memory, guardrails).

• Own the most complex, highest-impact agents end-to-end — from problem framing and prototyping through evaluation, deployment, monitoring, and continuous improvement in production at scale.
• Define the evaluation discipline: ground-truth datasets, offline and online evals, regression suites, replay tooling, and observability for non-deterministic agent behavior.
• Work directly with frontier models (Claude, GPT, Gemini) — benchmark them on security tasks and architect around their strengths and failure modes.
• Build shared patterns, frameworks, and infrastructure (prompting, tool use, retrieval, guardrails, cost controls) to reduce the cost of shipping the next agent.
• Partner with security and platform leadership to identify high-leverage opportunities and translate ambiguous problems into robust agent designs.
• Mentor and level up engineers through design reviews, pairing, and by setting standards for agent quality, safety, and reliability.
• Stay ahead of the field — track research, model releases, and emerging techniques, and convert relevant insights into a concrete roadmap.