Skip to Content
View all events

Agentic Coding with OpenCode

Published by O'Reilly Media, Inc.

Intermediate content levelIntermediate

AI-empowered development with the fastest-growing coding agent

What you’ll learn and how you can apply it

  • Learn to use OpenCode CLI for software development, maintenance, and more
  • Flexibly work with different LLM providers as needed
  • Practice strategies that work with coding agents, both for greenfield development and maintenance of mature applications
  • Avoid the traps and pitfalls that lead to AI regret

Course description

If you’re new to agentic coding and are not sure where to begin, software engineer and trainer Aaron Maxwell will show you how to boost your AI-enhanced development skills using OpenCode, the fastest-growing AI coding agent.

In this three-hour course, you’ll learn how to use OpenCode to create new applications, add features to mature applications, work with your unit test suite (or add one if the suite is missing), and quickly understand baffling code bases, all with an emphasis on sound engineering principles and best practices. Much of what you’ll learn transfers to Claude Code, OpenAI Codex, Gemini CLI, and dozens of other tool stacks.

Instead of feeling overwhelmed by AI, you’ll gain a clear understanding of how to use coding agents in real-world production software development tasks without creating future problems or technical debt. Plus, OpenCode supports nearly every LLM provider, so you’ll likely be able to use it with your current subscription.

This live event is for you because...

  • You’re a software developer, data professional, or DevOps engineer who wants to become an AI-assisted software developer.

Prerequisites

  • OpenCode installed on your computer prior to the course
  • Your LLM provider connected to OpenCode
  • Basic Python knowledge (most of the examples will be in Python)
  • Basic Git knowledge
  • Comfort using the command line of your OS (MacOS, Windows PowerShell, Linux)
  • Recommended: A cheap subscription to either OpenCode Go or OpenAI (Plus or better)

Schedule

The time frames are only estimates and may vary according to how the class is progressing.

Introduction to AI coding agents and OpenCode (30 minutes)

  • Presentation: Understanding agentic AI coding tools and their capabilities; Benefits of OpenCode compared to other coding agents; how to choose an LLM; initial setup of OpenCode
  • Hands-on exercise: Building a simple system statistics tool

Working with OpenCode (40 minutes)

  • Presentation: Plan mode vs. Build Mode; Prompt sequencing strategies; The relationship of LLM prompts to spec-driven development; exploring the demo application
  • Q&A
  • Break

Coding agent guidance and building your first application (40 minutes)

  • Presentation: Session management; Global and project-specific rules files; Guidance for AGENTS.md, with examples
  • Hands-on exercise: System monitoring TUI with AGENTS.md

Intro to Agent Skills (30 minutes)

  • Presentation: The shortcomings of AGENTS.md; how agent skills let you capture context-specific rules, instructions and resources; what agent skills are good for (and what they’re not for)
  • Q&A
  • Break

Building Agent Skills (40 minutes)

  • Presentation: Examples of useful agent skills; How coding agents decide which skills to use; Creating your own agent skills; Agent skill traps and how to avoid them
  • Hands-on exercise: Create an agent skill
  • Q&A

Mature Codebase Strategies (25 Min)

  • Presentation: Adding automated tests; Refactoring code; Modernizing code
  • Final Q&A

Your Instructor

Aaron Maxwell

Aaron Maxwell is author of the book "Powerful Python: The Most Impactful Patterns, Features, and Development Strategies Modern Python Provides." As a software engineer, he has worked in devops, test automation, and machine learning, and now divides his time between coding, writing, and teaching

Xlinksearch

Skills covered

  • AI & ML
  • AI Agents