Each skill is a few short lessons, then practice. You rate yourself at the start and again at the end.
Core to the job
Chapter 2
Security fundamentals
Spot phishing emails and fix a system's weak security settings
What you'll learn
CIA triad
Least privilege
Defense in depth
Attack surface
Practice
Triage 5 vulnerability scan findings25 min
Chapter 2
Threat hunting
Map attacker behaviour to logs, search for what alerts missed
What you'll learn
MITRE ATT&CK
Hunting hypothesis
Detection coverage
False positive
Practice
Map an Attack Technique to MITRE ATT&CK15 min
Chapter 4
AI feature safety
Spot tricks that make an AI ignore its rules
What you'll learn
Prompt injection
AI attack surface
Guardrail layers
Guardrail trade-off
AI threat model
Agent action risk
Practice
Spot Prompt Injection in Five Real Examples15 min
Craft a prompt-injection attack and write a guardrail30 min
Chapter 4
Machine learning basics
Trace an AI request and name where attackers could interfere
What you'll learn
Training and inference
Embedding vectors
Training data poisoning
Model attack surface
Practice
Match Six ML Terms to What They Mean for an Attacker15 min
Chapter 4
AI language models
Measure the size of AI requests and compare randomness settings
What you'll learn
Tokens and context
Sampling variation
Model trade-offs
Embeddings for search
Practice
Map a Chatbot's Context Window and Mark What an Attacker Controls20 min
7 more in the plan
Secure codingFind risks in a login process and remove exposed passwordsCh 2
What you'll learn
Trust boundaries
Access control
Secret handling
Dependency risk
Incident responseBuild an incident timeline from logs and write plain updatesCh 6
What you'll learn
User impact first
Mitigate before root cause
Blameless post-mortem
Incident command roles
Practice
Build a Timeline from Outage Logs and Find the Cause20 min
Write an on-call handoff note15 min
ScriptingWrite a script that checks disk space and warns youCh 6
What you'll learn
Script arguments
Error handling
Input checks
Readable scripts
Problem solvingTest each possible cause of a problem instead of guessingCh 7
What you'll learn
Problem vs symptom
Clean problem splits
Root cause
Deciding under uncertainty
Practice
Break an Ambiguous Problem into Actionable Parts20 min
Critical thinkingSpot a weak assumption and ask where numbers come fromCh 7
What you'll learn
The hidden assumption
Argue the other side
The base rate
Fact vs inference
Wanting it to be true
Practice
Spot the Weak Link in an Argument10 min
CommunicationRewrite an update for two different readers, cutting extra linesCh 8
What you'll learn
Lead with the point
Match detail to the reader
Said versus heard
Specific feedback
Practice
Write a Six-Line Delay Message10 min
Working with othersContribute to group work, ask for help and credit teammatesCh 8
What you'll learn
Early team friction
Clear ownership
Challenge ideas, not people
Lead or follow
Practice
Write a Request That Makes It Easy to Say Yes10 min
Your plan
Chapter by chapter.
12 chapters, in this order. Chapter 1 is free.
1~2 wks
See the work of an AI security engineer
StartFree
2~2 wks
Secure code and hunt for hidden threats
Skills
3~2 wks
Scan for weaknesses and fix the worst
Project
4~2 wks
Find where AI features break unsafely
Skills
51–2 wks
Talk to two working AI security engineers
People
6~2 wks
Handle incidents and automate repeat work
Skills
7~2 wks
Find the cause before you fix it
Skills
8~2 wks
Make yourself understood across a team
Skills
9~2 wks
Interview three working infrastructure professionals
Project
101–2 wks
Answer the questions interviewers ask
Interview
111–2 wks
Decide how you will get in
Decide
12on your timeline
Send your first applications
Apply
The job
What a AI Security Engineer does
Keeps machine-learning systems and AI agents safe from attack by finding weak spots, setting guardrails, and testing them before launch. Sits between security engineering and machine learning, so models behave the way teams expect.
The work
Map the AI attack surface across models, data, prompts, and agents
Defend against prompt injection, where bad input hijacks a model's behavior
Stop model theft and extraction so the private settings a model learned stay protected
Catch data poisoning, where attackers slip bad examples into training data
Set guardrails that block jailbreaks and unsafe model outputs
Red-team agents by attacking them on purpose before they ship
Review AI features with engineers and agree on security requirements
A typical day
Review alerts and logs from AI systems, then triage anything suspicious
Try fresh jailbreaks and prompt attacks against a model in a test setup
Meet machine-learning engineers to harden a new agent before it ships
Tune guardrails and content filters, then re-test that they hold
Write up findings and brief teams on the risks you found
How long does it take to become a ai security engineer with Welica?
The plan is 12 chapters, 4–5 months at 15 to 30 minutes a day. You can go faster or slower; the plan moves with you.
Do I need a degree?
Not always. Common routes are bachelor degree in computer science, cybersecurity, or machine learning, diploma or certificate in IT security plus self-study in machine learning, security bootcamp combined with online machine-learning courses and labs, or self-study path using AI security courses, home labs, and a portfolio.
What is free?
Chapter 1, See the work of an AI security engineer, is free for good. Premium unlocks the rest of the plan.