• Shape technical direction and lead ambiguous, cross-team Agentic SOC capabilities from concept through measurable production improvement.
• Design and build reusable capabilities spanning agent runtime, orchestration, governed tool use, context and memory, evaluation systems, trace analysis, and quality gates.
• Translate SOC, SIEM, detection, incident, investigation, and response needs into shared mechanisms in partnership with security-domain experts.
• Design agents and supporting systems for grounded reasoning, planning, delegation, human oversight, traceability, auditability, and safe failure handling.
• Define measurable success criteria for AI capabilities, including task completion, groundedness, tool-use correctness, trajectory quality, human intervention, safety, latency, cost, and reliability.
• Establish closed-loop, metrics-driven development using golden datasets, offline and online evaluations, production traces, SME and customer feedback, experiments, and regression gates.
• Connect AI quality metrics to production behavior and customer outcomes, identifying signals that are incomplete, misleading, or vulnerable to optimization without meaningful improvement.
• Architect for public cloud, on-premises, sovereign cloud, and air-gapped environments, addressing data residency, restricted connectivity, local model and tool availability, packaging, upgrades, and resource constraints.
• Establish patterns for identity propagation, authorization, tenant isolation, policy enforcement, privacy, auditability, and permission-aware actions.
• Evaluate, integrate, and drive adoption of technologies across Splunk and Cisco, contributing Security requirements and reusable improvements.
• Leverage AI tools throughout requirements, design, coding, review, testing, documentation, debugging, release, and production improvement, with appropriate human validation and security, privacy, and intellectual-property controls.
• Diagnose failures spanning models, prompts, tools, data, orchestration, services, infrastructure, and product integrations.
• Remain hands-on in implementation and design reviews, raise engineering standards, and mentor and influence senior engineers across teams.