My journey started with rooms, devices, identity, and employee support. Today I build production-minded AI agents shaped by the constraints of real IT operations.
My path started with the infrastructure people rely on every day—rooms, devices, identity, and support—and evolved into building AI systems that make those operations faster, clearer, and more repeatable.
Break into technology without relying on a traditional pathway or an existing professional network.
Technical skill creates momentum; communication, reliability, and coachability turn it into opportunity.
Enterprise IT readiness and a direct pathway into the Palo Alto Networks internship.
Keep more than 300 collaboration rooms reliable while working inside enterprise incident and asset processes.
Strong support scales through repeatable documentation, not one-off fixes that disappear with the technician.
AV and IT operations, service management discipline, and documentation for distributed teams.
Keep employees productive across endpoints, mobile devices, networks, and meeting technology in a high-expectation environment.
The technical fix and the employee experience are one outcome; clarity and trust matter as much as speed.
Endpoint operations, device management, compliance, and omnichannel support centered on the user.
Scale IT knowledge and decisions without losing accuracy, context, ownership, or human control.
Agents become valuable when grounded in real operational evidence and designed around explicit human boundaries.
Production-minded agent systems for support, knowledge management, asset analysis, and governed operations.
Manual research and drafting avoided across roughly 290 escalated CoPilot cases.
Estimated weekly documentation time saved through scheduled, governed runbook drafting.
Mac inventory records processed for fleet analysis.
Mac and Windows devices surfaced in a verified site dry run.
Four systems, each shaped around a real operational constraint. Open a profile for the design story—or run the workflow demo to see the sequence.
The common thread isn’t a model. It’s a way of building systems that can explain what they know, respect the boundary of what they can do, and get better from real work.
Solved tickets, internal runbooks, vendor knowledge, and system-of-record data—not generic intuition.
Separate high-confidence automation from low-confidence guidance, and keep a human at consequential gates.
Failures become sharper rules, stronger skills, better retrieval, and safer operating boundaries.