AI-first development, safely

A practical introduction to security when using AI tools in development. Together we identify risks in the tools you use every day and explore ways to handle data that do not slow the team down while still meaningfully reducing the chance of sensitive data leaking.

You will learn:

• The most common threats developers face in the AI era
• Risks and reality of AI tools: where your data goes, what AI tools can see, shadow AI
• Data classification: which data AI may work with, what may leave the corporate perimeter, and what must not
• When to use local models, RAG, and vector databases
• Context engineering basics: how to give AI the right context safely

We continue with a more hands-on segment focused on how you work during delivery.

You will learn:

• How to set up an AI workflow: skills, agents, context, MCPs…
• Which tools to choose and what to base decisions on
• How to reduce bad outputs, hallucinations, and generated vulnerabilities
• Code review in the AI era
• Guardrails and automated tests: locally and in CI/CD pipelines
• Safe handling of secrets: SSH keys, API keys, connection strings…

Once engineering is under control, it is time to think about how to take AI usage further and scale the opportunities it unlocks.

You will learn:

• Principles of AI agents, their orchestration, and building MCPs
• Risks of agentic systems
• Types of attacks on AI and how to prevent them
• Sharing practices between team members
• Spreading proven principles across teams