Trailwise: A Local AI Nudge to Touch Grass

Description and use

Trailwise is a local-first micro‑adventure planner that instantly generates short outdoor plans using an open‑weight AI model running in the browser, reducing decision friction and encouraging users to go outside.

Problem

Users struggle to decide on quick outdoor activities and often keep their plans on a screen instead of going outside.

Outcome

Trailwise enables instant, privacy‑preserving micro‑adventures by running an open‑weight model locally, reducing decision friction and encouraging outdoor activity.

Workflow

• 1. User enters location, available time, energy level, and preferred mood.
• 2. Browser app sends input to local FLAN‑T5‑small model via Transformers.js.
• 3. Model generates a short, friendly introduction for the outdoor plan.
• 4. If model download is pending or device is low‑powered, deterministic plan library provides a fallback plan.
• 5. User follows the generated or fallback plan and goes outdoors.

Evidence

Trailwise is a local‑first micro‑adventure planner that helps people spend less time deciding what to do and more time outside.

Source

The AI layer uses Xenova/flan-t5-small, an open‑weight model, through Transformers.js. The model runs directly in the user's browser and generates a short, friendly introduction for the outdoor plan.

Source

The app also includes a small deterministic plan library. This keeps the core experience useful while the model is downloading and gives lower‑powered devices a reliable fallback.

Source