AI-Agents-Projects-Tutorials

Why we included this project

Each entry pairs a Marktechpost article with a runnable notebook or Python script, making this a hands-on companion to the site's tutorials. Instead of only reading about agentic design, you can follow working code that shows how tool calling, session memory, and coordination between agents fit together in practice. That breadth is the main draw: one project demonstrates a Kimi CLI coding workflow, another builds a policy-governed financial research agent with Omnigent, and a third walks through skill-driven Claude agents with MCP connectors, so comparing how each approach handles the same problems takes little effort. The notebooks run in Google Colab, which means a small team can test a pattern without provisioning infrastructure first. This is a curated learning index rather than deployable software, but the code-backed recipes spare you most of the trial and error that comes with adopting agentic techniques.