Building AI Coding Assistant from scratch

Designed for forward-thinking software engineers, AI developers, and system architects, this course offers an immersive, hands-on journey into the inner workings of next-generation development tools. If you have ever wondered how state-of-the-art coding assistants like Cursor, Claude Code, and GitHub Copilot actually interpret codebases, suggest refactors, and execute terminal commands, this program provides the definitive blueprint. Under the guidance of Alexsandro Souza, you will move far beyond basic API calls to construct your very own autonomous coding agents from scratch. The curriculum dives deep into the architecture of modern LLM agents, focusing on the powerful ReAct (Reasoning and Acting) framework that enables models to think step-by-step and interact with external file systems. You will learn how to orchestrate complex workflows using premier agentic libraries such as LangChain, CrewAI, and LlamaIndex. Additionally, the course emphasizes advanced prompt engineering and context retrieval strategies, ensuring your custom assistants produce highly accurate, syntactically correct code. By the end of this course, you will possess the specialized skills to build, customize, and deploy intelligent, self-correcting coding agents that adapt to real-world developer workflows.