Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

Video course · 15 chapters · 134 min · certificate

Agentic AI in Python: Build a Coding Agent with Google Gemini API

Build a coding agent in Python with the Gemini API: what agents are, four file tools, project setup with uv, API calls, CLI args and history, a verbose flag, building get-files, read-file, write-file and run-python tools, security, the system prompt, function declarations, function calling and the agent loop.

Become a memberAlready a member? Sign in

What you'll learn

  • Explain what makes an AI agent
  • Set up a Python project and call the Gemini API
  • Build file and execution tools
  • Declare functions for the LLM
  • Implement function calling and an agent loop
  • Understand the security risks of agent-run code

Chapters

15 chapters · 134:23
  1. 6:03 01Intro Members Why build an agent Course overview and what an agent is.
  2. 9:41 02Setup Members Four tools and setup The four tools, prerequisites and uv setup.
  3. 9:00 03API Members Gemini API basics Making the first API call.
  4. 5:55 04CLI Members CLI arguments and history Accepting prompts and tracking conversation.
  5. 5:44 05Debug Members Verbose flag and test project A verbose flag and a calculator test project.
  6. 13:16 06Tools Members Tool 1: get files info Listing a directory safely.
  7. 8:45 07Tools Members Tool 2: get file content Reading files with truncation.
  8. 9:06 08Tools Members Tool 3: write file and security Writing files and the dangers of running AI code.
  9. 10:30 09Tools Members Tool 4: run Python file Executing Python with subprocess.
  10. 15:10 10Prompting Members The system prompt Guiding the agent's behaviour.
  11. 8:28 11Tool calling Members Declaring functions How LLMs call functions via schemas.
  12. 7:41 12Tool calling Members All declarations Adding schemas for every tool.
  13. 8:11 13Tool calling Members Function calling logic Executing the model's function calls.
  14. 9:41 14Agent Members The agentic loop Looping until the task is done.
  15. 7:12 15Demo Members Final demo: fixing a bug The agent fixes a bug on its own.