Identity security automation
In production
Automated employee offboarding pipeline
An offboarding ticket now triggers the full identity teardown automatically: the Entra ID
account is disabled, active sessions are revoked, licenses are reclaimed, and the outcome
is written back to the ticket as an audit trail. Closes the window where a departed
employee still holds a valid token — the gap manual offboarding always leaves open.
n8nMicrosoft GraphEntra IDWebhooks
Incident response
Resolved
Log4Shell alert investigation (CVE-2021-44228)
An endpoint security alert flagged possible Log4Shell exploitation. I wrote a PowerShell
investigation script, deployed it through RMM to enumerate Java runtimes and Log4j
artifacts, and established the host had neither — it was not exploitable. Traced the
flagged source address to the network's own DNS gateway. Documented as detection of
scanning traffic, not a compromise.
PowerShellEndpoint securityThreat analysisForensics
Endpoint security
Deployed
Fleet-wide endpoint agent deployment
Silent install and upgrade tooling for endpoint protection across a distributed fleet,
hardened through repeated real-world failure modes — partial installs, stale agents, and
hosts where the installer ran without a user context. Turned an unreliable manual rollout
into a repeatable one.
PowerShellRMMEndpoint hardening
Identity auditing
Ongoing
Entra ID audit and investigation tooling
A growing PowerShell toolkit for identity questions the portal answers slowly or not at
all: reconciling directory users against authoritative HR data, resolving Entra object IDs
from Windows security events during local administrator group investigations, and
reporting license and service plan state by stable GUID rather than fragile display names.
PowerShellMicrosoft GraphEntra IDEvent logs
Software engineering
Shipped
Internal asset and logistics platforms
Two internal web applications built end to end: an asset management system in Flask for
hardware lifecycle tracking, and a shipment tracking portal in Node.js and Express with
JWT authentication, four-tier role-based access control, carrier API integration, and
activity logging throughout — backed by a 200+ test suite. Security engineers who can read
and write real applications make better decisions about them.
Node.jsExpressFlaskJWTRBAC
Detection engineering
Ongoing
Detection engineering lab
A hybrid lab — on-premises Active Directory synced to Entra ID, endpoints reporting into
Sentinel and Defender — used to run attack techniques, observe the telemetry they generate,
and write detections against them. Every technique ends with a tested rule and a write-up
on the blog.
Microsoft SentinelKQLAtomic Red TeamSysmonSigma