Turning security evidence collection into a repeatable workflow
A PowerShell-based system transformed a multi-hour, ten-step evidence-gathering process into an on-demand workflow.
AI-native IT operations · systems · identity
I design calmer identity, endpoint, SaaS, and security operations for growing teams—using thoughtful automation, clear documentation, and AI-assisted tools where they genuinely help.
What I do
The same pattern across all of them: understand the real workflow, secure it by default, automate the repeatable parts, and document the rest.
Okta, SAML/OIDC/SCIM, MFA, Device Trust, RBAC, Entra ID and Active Directory.
PowerShell, Python, Bash, REST APIs, Okta Workflows, n8n and GitLab CI/CD.
Kandji, Jamf, macOS, Google Workspace, Microsoft 365, AWS WorkSpaces, GCP.
Incident response, SLA management, root cause analysis, ITIL and PagerDuty.
Selected work
A PowerShell-based system transformed a multi-hour, ten-step evidence-gathering process into an on-demand workflow.
Slack, Okta Workflows, Zapier, and Notion worked together to reduce coordination and documentation overhead during defined incident scenarios.
Okta Device Trust, Kandji, and access-request workflows created a more secure and understandable path to employee access.
The operating pattern
My best work sits where people, security, and systems meet. I look for repeated effort, unclear ownership, and brittle handoffs—then turn them into understandable, reusable operations.
Map the real process, including the awkward exceptions.
Secure defaults, visible decisions, and a human fallback.
Documentation and observability are part of the system.
Field notes
A portfolio can show polished outcomes and still expose the learning process that made them possible.
Read note →If an automation only works when its author is present, the automation is still carrying hidden manual work.
Read note →Now
CURRENT EXPERIMENT
Building kellray.com as a public résumé, a writing practice, and a safe place to test the tools I’m learning.
Open the build note →