Investigation & Prompt Engineering Masterclass
Solution overview
Everything you've built leads to this.
You have 9 skills. 6 data sources. An AI analyst that knows your network. And a live environment with real security alerts firing right now.
In Part 1, you investigate — with no script to follow. Real alerts on the Domain Controller. Real telemetry across every data source. You decide what to ask, what to dig into, what to correlate. Nobody tells you the answer because there isn't one right answer — there's your investigation and the evidence you find.
In Part 2, you witness the moment that changes how you think about AI in security. Two prompts. Same model. Same data. Opposite conclusions — "monitor for 24 hours" versus "isolate the Domain Controller now." Then you write your own engineered prompt and assess whether it produced the analysis you intended.
You leave with the 5 principles that make every AI interaction sharper — not just in security, but anywhere you use AI to analyze data.