My First Switch Hackathon: Setting Up an AI-Powered Workspace

Introduction

For the Flip the Switch Hackathon, I decided to explore how an AI-powered workspace could help organise and manage creative tasks.

My goal was to understand how Switch works, set up a local environment, and explore connecting an AI coding agent through OpenCode.

Setting Up Switch

I started by setting up Switch on my Windows computer. The process introduced me to several tools and concepts:

• Switch: The workspace I wanted to explore.
• Docker: Used as part of the local environment.
• PostgreSQL: The database used by the setup.
• OpenCode: The AI coding agent I explored connecting to Switch.

Getting all these pieces installed and running took patience. I had to understand how the different tools fit together rather than treating setup as one single step.

Exploring OpenCode

I initially had an older OpenCode CLI version installed. I then downloaded a newer standalone version and checked that it worked from PowerShell.

I also checked the available models and confirmed that OpenCode could respond to a simple test prompt.

This helped me understand the difference between installing a tool, authenticating it, and connecting it to another application.

What Worked and What Didn't

I successfully got the local Switch server running and confirmed that the OpenCode CLI could run independently.

However, I ran into a problem when Switch tried to verify the OpenCode provider sign-in. The provider verification did not complete, so I could not finish creating and connecting my intended agent.

I decided not to pretend the integration was successful. Instead, I treated this as an important part of the learning experience.

What I Learned

This setup taught me a few useful lessons:

1. A running local server does not automatically mean every integration is working.
2. Version differences between tools can create confusing setup problems.
3. Authentication and provider verification are separate from simply installing a CLI.
4. Reading logs and testing one component at a time makes troubleshooting easier.
5. It is important to document failures honestly, not just successful results.

Conclusion

My first experience with Switch was a learning journey through local development, databases, command-line tools, and AI agent integration.

I did not complete the final provider connection, but I gained a better understanding of the setup process and the challenges involved in connecting independent tools.

This experience has given me a starting point to keep experimenting with AI-powered workflows and build something more complete in the future.

Created as my learning reflection for the Flip the Switch Hackathon.