This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend
What I Built
I built Sidequest, a local-first hangout planner for the friend who wants to spend time together but could use a little help making the plan.
Share their interests, location, mood, budget, and how much time you have. Sidequest creates a thoughtful, personalized outing with a few activities and a small extra touch.
Demo
Run Sidequest locally by following the setup instructions in the repository README. Once the server is running, open http://127.0.0.1:8000.
Code
View the Sidequest GitHub repository
How I Built It
Sidequest uses the open-weight Qwen 2.5 3B model through Ollama for local inference. The frontend uses HTML, CSS, and JavaScript, with a lightweight Python server connecting the app to Ollama.
The friend’s details stay on the device: the browser sends them to the local server, which passes the prompt to the local model. No hosted AI API or account is needed.
Why Does Open Innovation Matter?
Open innovation makes local, private AI applications like Sidequest possible. Running an open-weight model on your own computer gives you control over personal details and lets you use the project without depending on a hosted AI service or API key. The code and model choice can also be inspected and adapted.
Top comments (1)
I feel good