This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass

What I Built
🌿 What is FIELDNOTE?
“The better you use FIELDNOTE, the less you need to look at your phone.”

FIELDNOTE is a local-AI observation engine designed for the Hacktoberfest 2026 Week 1 “Touch Grass” Challenge. Rather than competing for your screen time with streaks, push notifications, and gamification, FIELDNOTE is an intentional physical instrument that prompts you to put your phone away, step outside, and observe the living world around you.

Demo

fieldnote-97c7.onrender.com

Code

rudrarajsinh920 / fieldnote

🌿 FIELDNOTE

“AI that teaches you to notice.”
A collectible field journal from the future for Hacktoberfest 2026 Week 1 “Touch Grass” Challenge.

🔗 Live Application URL: https://fieldnote-97c7.onrender.com/

---

Core Product Philosophy

“The better you use FIELDNOTE, the less you need to look at your phone.”

FIELDNOTE is not another outdoor GPS tracker, gamified streak app, or generic conversational AI chatbot. It is a local-AI observation engine designed to minimize screen time and maximize physical sensory presence.

The Naturalist Ritual Loop

PREPARE (Define sensory window & intention)
   ↓
FIELD INVESTIGATION (Local AI generates an observation directive)
   ↓
FIELD MODE (7 seconds of inactivity → interface quietly disappears)
   ↓
PHONE AWAY (Observe physical reality with your senses)
   ↓
RETURN (Record raw notes, optional specimen photo & voice note)
   ↓
NATURALIST REFLECTION (Local AI distills ecological dynamics & inquiry)
   ↓
THE LOGBOOK (Permanently sealed into collectible on-device journal)

---

System Architecture

FIELDNOTE is architected…

View on GitHub

How I Built It
🔄 The Naturalist Ritual Loop
Prepare: Select a sensory lens (School & Science, Community & Environment, or Everyday Discovery) and define your outdoor observation window (e.g., 10 minutes).
Investigation Directive: Local open-weight AI (Ollama llama3.2:3b) synthesizes a grounded sensory prompt tailored to noticing patterns, textures, lichen symbiosis, leaf venation, or micro-habitats.
Field Mode (Screen Disappears): After 7 seconds of inactivity, the digital UI quietly dims and disappears, freeing your attention to explore nature with your actual senses.
Physical Observation: Explore the wilderness, park, or urban pavement with your phone in your pocket.
Return & Capture: Return to record sensory notes, capture macro specimen photography, or dictation recordings.
Naturalist Synthesis: Local AI distills observations into ecological dynamics, distinguishing verifiable evidence from hypothesis.
Collectible Logbook: Permanently sealed into an on-device naturalist folio with SQLite persistence and offline PWA capabilities.
🛠️ Technical Stack & Architecture
Frontend (@fieldnote/client): React 19, TypeScript, Vite, Tailwind CSS, Lucide icons, responsive from 375px mobile to 1440px+ desktop.
PWA & Offline Capability: Service Worker caching via Workbox for complete offline fieldwork in remote areas without signal.
Backend (@fieldnote/server): Node.js 24, Express, TypeScript, native node:sqlite database with Write-Ahead Logging (WAL).
Local AI Engine: Open-weight inference via Ollama (llama3.2:3b). 100% private, on-device computation with zero cloud tracking or telemetry.
Test Suite: Automated test suite with 10/10 passing tests validating AI recovery, timeout handling, state machine invariants, and SQLite storage.
Design System: Botanical editorial aesthetic inspired by vintage field guides, warm ivory paper tones (#FCFBF7), deep forest green (#12372A), and high-resolution landscape artwork.
Why Does Open Innovation Matter?

My Agent Session

Prize Categories