What Can We Really Do with Claude Code and OpenAI Codex?

A lot changed in AI during the last year.
Before, many discussions were about prompts, chatbots and which model is better. Now the important change is different: tools like Claude Code and OpenAI Codex can work with files, code, commands, documentation, reports and real tasks.

They are called coding agents, but this does not mean they are only for developers. In real work they can help with coding, Excel and reporting tasks, documentation, DevOps, security checks, small internal tools, and many other structured tasks.

In this AI Talks session, Niko Panchev and myself will show practical examples of what people can do with Claude Code and OpenAI Codex today.
We will also discuss what this means for companies. If people start using AI agents for real work, companies need to think about data, security, AI literacy, governance and EU AI Act readiness.

Finally, we will show how AAE Framework can help organizations use agentic engineering in a more controlled/ enterprise ready way - with specs, reviews, tests, quality gates, documentation and delivery evidence.

As usual this will be hosted in our channel in YouTube on 21st of May 10:00AM Sofia Time.

**https://www.youtube.com/@AITalksBulgaria/videos**

The session is for everyone who wants to understand how AI agents can help with real work.

See you there!