Skip to content

Course

Detection Engineering

Build detections that fire on attackers, not noise: the telemetry that matters, Sigma-style rule craft, ATT&CK mapping that stays honest, false-positive tuning, atomic tests, and real coverage metrics.

9 lessons

Course Q&A

Telemetry: You Cannot Detect What You Do Not Collect

Before you write a single rule, get the data. Process creation, command lines, Sysmon, and a pipeline you can actually query.

  1. 01Choosing Telemetry That Sees AttackersAV alerts and firewall logs will not catch a living-off-the-land intruder. Here is the telemetry that actually does.20mOpen
  2. 02Command Lines and Process Trees: Where the Truth LivesModern malware runs signed Microsoft binaries. The maliciousness lives in the arguments and the parent process.25mOpen
  3. 03A Log Pipeline You Can Actually QueryWEF, forwarders, and field normalization — getting Sysmon from fifty endpoints into one searchable place.30mMembers

Rule Craft: Detections That Survive Contact

Sigma anatomy, behavior-based logic that outlives IoCs, and MITRE ATT&CK mapping that does not inflate your coverage.

  1. 01Anatomy of a Sigma RuleThe vendor-neutral detection format: logsource, selections, conditions, modifiers, and converting to your SIEM.30mMembers
  2. 02Detect Behavior, Not StringsIoCs rot in days; parent-child relationships and abuse patterns last years. Write rules on what attackers must do.25mMembers
  3. 03Mapping to MITRE ATT&CK Without LyingTechnique vs sub-technique, tagging what you actually detect, and why a rule is not coverage.20mMembers

Prove It: Testing, Tuning, and Honest Coverage

Tame false positives without gutting the signal, fire atomic tests to prove rules work, and measure coverage you can defend.

  1. 01Tuning False Positives Without Deleting the SignalBaselining, surgical filters, and the exclusion-list trap — how to quiet a rule while keeping the malicious case.30mPremium
  2. 02Atomic Tests: Prove Your Detection FiresAn untested rule is a hypothesis. Atomic Red Team, the fix loop, and negative testing.25mPremium
  3. 03Measuring Coverage Without Fooling YourselfTelemetry, rule, and validated layers; alert fidelity and MTTD; and why '80% ATT&CK coverage' is usually a lie.25mPremium