Back to Event Recordings
Workshop Resource

Building a Coding Agent: Python/Django Edition

Build your own project bootstrapper: a coding agent similar to Lovable, but for Python/Django projects. Learn how to create a tool-using agent that can bootstrap complete applications from natural language instructions.

August 14, 2025Intermediate to Advanced
ai-agentscoding-agentsdjangopythonllm-engineering

Timestamps

Click any timestamp to jump to that moment in the video

00:00
Introduction & Workshop Goals

Overview of the workshop objectives and what we'll build

00:09
Presenter Background & Experience

Introduction to the presenter's background in AI and coding agents

01:08
Example: What is 'Lovable'? (The Goal)

Understanding Lovable as inspiration for building our own coding agent

02:18
The Plan: Building a Python/Django Coding Agent

Outlining the approach to build a coding agent for Python/Django projects

03:43
Tools Setup: OpenAI API Keys & Budget

Setting up OpenAI API keys and configuring budget limits

05:02
Environment: Using GitHub Codespaces

Setting up the development environment with GitHub Codespaces

11:20
Concept: Agents vs. Chatbots (Adding Tools)

Understanding the difference between chatbots and agents, and how tools enable agentic behavior

16:48
Defining System Prompts & User Prompts

Crafting effective system prompts and user prompts for the coding agent

24:00
Setting up the Django Project Template

Preparing a Django project template as the base for the agent to work with

29:01
Defining Agent Tools (Read/Write Files, Search)

Implementing core tools: file reading, writing, and search capabilities

33:36
Implementing & Testing Tools in Jupyter

Building and testing the agent tools interactively in Jupyter notebooks

37:27
Crafting the 'Developer Prompt' (Crucial Step)

Creating the critical developer prompt that guides the agent's behavior

42:16
Running the Agent with ToyAIKit

Executing the coding agent using ToyAIKit framework

47:49
Recap & Break

Reviewing what we've built so far and taking a break

50:42
Intro to OpenAI Agents SDK

Introduction to OpenAI's Agents SDK for building production agents

53:48
Building a Joke Agent with Agents SDK

Creating a simple joke agent to understand Agents SDK patterns

58:08
Porting the Coding Agent to Agents SDK

Migrating our Django coding agent to use OpenAI Agents SDK

1:04:07
Intro to PydanticAI (Production Grade Framework)

Introduction to PydanticAI as a production-ready framework for agents

1:06:50
Creating the Agent with PydanticAI

Rebuilding the coding agent using PydanticAI framework

1:10:56
Switching to Anthropic Claude 3.5 Sonnet (Better Code)

Comparing models and switching to Claude for better code generation

1:15:34
Test Case: Building an Anki Cards App

Testing the agent by building a complete Anki cards application

1:16:16
Testing Z.AI Models (Reasoning Models)

Exploring reasoning models from Z.AI for advanced agent capabilities

1:25:45
Course Overview & Learning Path

Overview of the AI Engineering Buildcamp course and learning journey

Core Tools

OpenAI APIOpenAI Agents SDKPydanticAIAnthropic Claude APIToyAIKitDjangoPythonJupyter NotebookGitHub CodespacesZ.AI Models

What You'll Learn

  • Understanding the difference between chatbots and agents
  • Building a project bootstrapper from scratch for Python/Django projects
  • Implementing agent tools (file operations, search)
  • Crafting effective system and developer prompts
  • Using ToyAIKit for interactive agent development
  • Migrating to OpenAI Agents SDK
  • Building production-ready agents with PydanticAI
  • Comparing different LLM providers (OpenAI, Anthropic, Z.AI)
  • Testing agents with real-world projects
  • Understanding how project bootstrappers work internally

Expected Outcome

A fully functional project bootstrapper: a coding agent that can build Django applications based on user instructions, implemented using multiple frameworks (ToyAIKit, OpenAI Agents SDK, PydanticAI), with the ability to read/write files, search code, and generate complete applications from scratch