↓ Skip to main content
still figuring out AI, one post at a time
  1. MindSecSet/

still figuring out AI, one post at a time

·597 words·3 mins·
MindSecSet
Author
Virtue of Vague
Table of Contents

a series for Security analysts learning AI from the ground up.

three tracks. one concept per post. short reads. built for people who work on top of AI systems every day and want to understand what’s underneath.

ONE CONCEPT AT A TIMEAI RED TEAMER — 12 POSTS + HTB LABSML, deep learning, LLMs — HTB pathCLAUDE 101 — 6 POSTSprompting, Projects, Research modeAI FLUENCY — 6 POSTSdelegation, description, discernment,diligenceno PhD required — just a Security analyst,one concept at a time.

AI Red Teamer
#

twelve posts covering ML/deep-learning fundamentals from the ground up, following Hack The Box’s own AI Red Teamer path — plus ongoing hands-on lab notes as I clear each certification module.

#post
0why a security analyst is learning AI (and why you should too)
1AI, ML, deep learning — same thing? not quite
2teaching machines with examples — like training a junior analyst
3your SOC playbook is literally a decision tree
4drawing the line — how machines separate threats from noise
5no labels, no problem — finding patterns in the chaos
6this is literally what your SIEM does — now let’s understand it
7learn by doing — how AI figures things out the hard way
8the brain analogy everyone uses — here’s what it actually means
9how AI sees images and reads sequences — two tools, one post
10the AI everyone is talking about — how does it actually work
11from noise to meaning — the quiet revolution in AI

Hack The Box lab notes
#

hands-on lab notes from Hack The Box’s AI Red Teamer certification path — building security-focused ML models, then learning where each one is vulnerable. raw, ongoing, updated as I clear modules.

#post
2HTB AI Red Teamer — Module 2: when your model’s blind spot is someone else’s attack surface
3HTB AI Red Teamer — Module 3: when your model is a liar, a thief, and a backdoor
4HTB AI Red Teamer — Module 4 Pt 1: how I tricked an AI into spilling its secrets
4HTB AI Red Teamer — Module 4 Pt 2: jailbreaks, defenses, and the chatbot that banned its own CEO

Claude 101
#

six posts on actually using Claude — prompting, Projects, connectors, Research mode, and more. built on Anthropic’s Claude 101 course, filtered through a Security analyst’s lens.

#post
1from open to actually useful — a Security analyst’s guide to Claude
2how to actually talk to Claude — and get something useful back
3stop starting from scratch — organising your work in Claude
4Claude meets your tools — connectors and enterprise search
5Research mode — when Claude stops answering and starts investigating
6Claude everywhere — desktop, code, slack, excel, chrome

AI Fluency
#

six posts on the AI Fluency framework — delegation, description, discernment, and diligence. built on Anthropic’s AI Fluency course, filtered through a Security analyst’s lens. plus a cheat sheet covering every term in the framework.

#post
1there is a difference between using AI and being fluent with it
2the AI fluency framework — a smarter way to think about working with AI
3not everything should go to AI — the art of delegation
4the gap between what you mean and what AI understands
5just because it sounds right doesn’t mean it is
6accountability doesn’t disappear when AI helps you
—AI fluency cheat sheet — key terms and concepts

Related