SOC analysts were blocked by repetitive work but had no way to automate it without engineering support.
<10 minutes to first AI agent; ≥80% analyst satisfaction in pilot.
Lead UX Designer — 0→1 product design, flows, AI trust patterns.
Design for the 3am analyst. Cognitive load compounds under pressure.
Netenrich's mission is to transform cybersecurity from reactive firefighting into adaptive, AI-driven operations. AgenZ is the agent builder that makes this real — enabling SOC analysts to create, test, and deploy AI agents that automate routine security tasks without writing a single line of code.
I led the end-to-end UX design, from early research through to shipped product, designing both the chat-based and form-based agent creation flows, live testing capabilities, and the usage analytics dashboard.
Through shadowing sessions and interviews across multiple SOC teams, four recurring pain points emerged that were limiting analyst productivity and threatening SLA compliance.
Analysts wasted hours on the same triage steps daily. SLA risk grew with every repeated manual action.
Only senior staff knew the right procedures. When they were unavailable, quality dropped sharply.
Different outcomes across shifts and timezones. No standardized playbook for common incident types.
Manual steps delayed incident resolution. Critical alerts sat in queues while analysts handled routine work.
AgenZ serves four distinct user groups, each with different needs and technical comfort levels.
We conducted shadowing sessions across shifts, interviewed analysts and managers, reviewed closure notes and escalation logs, and mapped the full detection-to-resolution workflow.
AgenZ offers two creation paths — chat-based for speed and form-based for governance — plus live testing, custom instructions, and usage analytics.
Natural language interface that lets analysts describe what they want in plain English. The AI interprets the intent, suggests a configuration, and builds the agent — all within a conversational flow.
Structured configuration for complex agents that need governance controls. Step-by-step wizard with validation, role-based permissions, and approval workflows.
Before deployment, analysts can test agents against real or simulated data. A split-pane view shows the agent's reasoning alongside its actions — building trust through transparency.
Agents can be configured with specific triggers (alert type, severity, source) and custom instructions that encode team-specific procedures and escalation rules.
Dashboard showing agent performance — tasks completed, time saved, error rates, and analyst feedback scores. This data drove adoption by proving ROI to skeptical teams.
