LW.
DE · SUBLIME
LW.
WRITING / NOTES

Detections, written like prose.

Field notes on detection engineering, machine learning, and the craft of finding what hides in the noise — plus short fiction away from the terminal.

WRITING / NOTES

Detections, written like prose.

Field notes on detection engineering, machine learning, and the craft of finding what hides in the noise. Most pieces are published on the Sublime Security blog.

SELECTED — RECENT
Hiding attacks in plain sight with invisible Unicode tag characters

How invisible Unicode tags hide adversarial instructions in email aimed at AI scanners and tools.

Sublime Security ↗
Calendar phishing attacks expanding across service providers

Research into malicious calendar invitations sent through legitimate service providers.

Sublime Security ↗
Surge in callback phishing attacks abusing auto notifications, verifications, alerts, receipts, and more

How attackers embed callback lures in automated notifications from trusted services.

Sublime Security ↗
Prompt injection attacks don’t look like what you’re seeing in social media and headlines

In email attacks, adversaries most often use prompt injection to confuse AI security analysis (for now).

Sublime Security ↗
Using AI signals within malicious email for attack detection and threat hunting

AI signals within email attacks can bolster detection, prevention, and threat hunting.

Sublime Security ↗
Forgetful Foes and Absentminded AIs

A cautionary tale of how I'm detecting and hunting for LLM generated emails, and some tools for you to do the same.

Read →
You’ve been invited to join a Meta for Business scam!

Credential phishing with fake and hijacked Meta for Business programs.

Sublime Security ↗

MEDIA & INTERVIEWS