Zulia Shavaeva← Back to Zulia
Free AI and security guide

Find a starting point in AI security

New to AI, building, leading or working with customers? Choose a learning route without a quiz →

Answer eight questions for an initial suggested path. Add optional skill details to refine it, or choose study hours to create a plan. The result helps you choose what to explore next; it does not measure job readiness.

Start the eight-question quiz See the 7 roles first
No signupFree guideWritten scoring rulesBased on a 48-posting research sample
The BuilderAI Security Engineer16 of 48 postingsThe BreakerAI Red Teamer9 of 48 postingsThe ArchitectAI Security Architect7 of 48 postingsThe GovernorAI Governance and Risk Lead7 of 48 postingsThe AssessorAI Compliance and Assurance Lead4 of 48 postingsThe AdvisorAI Security Consultant5 of 48 postingsThe ResearcherAI Security Researcher4 of 48 postings

Seven paths

Explore the work each role involves, follow the learning steps and try a practice project. The paths draw on a small research sample of job postings.

AI Security Engineer

Build and test security controls for applications that use language models and agents.

16 of 48 postings. Learn first:
LLM security: prompt injection, jailbreaks, output handling 94%Agent and agentic system security: tool use, permissions, memory, multi-agent 81%Python and coding for AI security work 81%

AI Red Teamer

Test AI applications for prompt injection, data leaks and unsafe tool use. Write findings that an engineering team can reproduce and fix.

9 of 48 postings. Learn first:
AI red teaming and adversarial testing 100%Python and coding for AI security work 89%LLM security: prompt injection, jailbreaks, output handling 78%

AI Security Architect

Design how AI systems handle data, call tools and enforce access boundaries.

7 of 48 postings. Learn first:
AI system threat modelling and secure architecture 86%LLM security: prompt injection, jailbreaks, output handling 100%Agent and agentic system security: tool use, permissions, memory, multi-agent 86%

AI Governance and Risk Lead

Assess AI risks and help teams decide which controls, reviews and evidence they need.

7 of 48 postings. Learn first:
AI governance, risk and compliance: NIST AI RMF, ISO/IEC 42001, EU AI Act, CSA AICM 86%LLM security: prompt injection, jailbreaks, output handling 100%AI system threat modelling and secure architecture 43%

AI Compliance and Assurance Lead

Assess AI systems against applicable obligations and prepare evidence for audits and assurance reviews.

4 of 48 postings. Learn first:
AI governance, risk and compliance: NIST AI RMF, ISO/IEC 42001, EU AI Act, CSA AICM 86%LLM security: prompt injection, jailbreaks, output handling 100%AI system threat modelling and secure architecture 43%

AI Security Consultant

Help client teams understand AI security risks and plan the work needed to address them.

5 of 48 postings. Learn first:
Advisory and customer-facing delivery skills 40%LLM security: prompt injection, jailbreaks, output handling 80%Securing model and agent workloads on cloud platforms 80%

AI Security Researcher

Investigate how AI systems fail, test attacks and defences, and publish reproducible results.

4 of 48 postings. Learn first:
AI red teaming and adversarial testing 100%Python and coding for AI security work 100%LLM security: prompt injection, jailbreaks, output handling 100%

How this was built

Research behind the guide

48 public job postings were read and coded for 14 skills by hand. Six clusters came from that coding. The seventh path, compliance and assurance, uses a subset of those postings. Small samples are labelled.

Free resources come first; paid options are labelled. Link status reflects stored checks, not the build date. 0 of 66 links could not be confirmed and are marked.

Limits of the guide

A prototype under validation. The diagnostic is v1.0 of a set of written rules, tuned against ten personas, not real outcomes. Pay visuals use separately sourced general-security benchmarks, not a forecast for AI roles or a personal salary estimate. Nothing here is a hiring promise.

The scoring rules are available to inspect. Course prices and access requirements can change; check the provider before enrolling.