Loading...
Loading...
Part of the bootcamp
Learn AI Agents from Scratch is one workshop inside AI Bootcamp for Software Engineers.
See the full path with mentor reviews and a portfolio outcome.
Frameworks hide the hard parts until they break. Build memory, tool calling, and autonomous loops from scratch with pure Python.
Engineers are learning here from
Frameworks hide the hard parts until they break. Build a real AI agent from scratch: memory, tool calling, autonomous loops, error recovery, and human-in-the-loop safety. Finish with a capstone Office Manager agent that ties it all together.
Build AI agents with pure Python. No LangChain. No CrewAI. No magic.
What you'll ship
What you'll learn
Curriculum
Agent fundamentals
Understand the difference between stateless LLMs and stateful agents. Build your first agent with conversation memory using pure Python.
ReAct pattern
Learn the Thought-Action-Observation reasoning loop that enables agents to use tools and solve multi-step problems.
Tool integration
Replace fragile regex parsing with native function calling using JSON schemas for reliable tool integration.
Agent loop
Build an autonomous agent that chains multiple tool calls in a while loop with safety limits.
Memory basics
Solve context window limits with sliding window and summarization memory strategies.
AI workflows
Build modular agent systems with the router pattern for task classification and the chaining pattern for multi-step workflows.
Error handling
Make agents resilient with validation, the reflection pattern for self-correction, and retry loops.
Human-in-the-Loop
Add safety gates and approval workflows so humans can review and approve sensitive agent actions.
Practical agent: capstone
Integrate everything you learned into a complete Office Manager agent with memory, tools, loops, error handling, and human-in-the-loop safety.
Who it's for
You want to ship AI features but frameworks feel like black boxes you can't debug.
You use LangChain daily but couldn't build an agent without it. That bothers you.
Your team is evaluating agent frameworks and you need to make an informed decision.
FAQ
Nope. We build from scratch. Pure Python + LiteLLM.
Anything that supports function calling. LiteLLM handles OpenAI, Anthropic, Google AI Studio (free), and more.
About 10 hours. Most people finish in 2-3 sessions over a weekend.
Yes. These are the same patterns running in production agent systems.
Build AI agents with pure Python. No LangChain. No CrewAI. No magic.