Skip to content
UsecaseBoard
← All use cases
dev.toGardening

Turn saved kitchen scraps into a printable gardening plan

At a glance

Second Harvest uses a local Gemma model to match leftovers with a beginner's growing space and supplies. This workflow is listed under Gardening, Food Waste, Local AI.

SourceBuilding a tool for these workflows? See weekly rates.

What they did

1) On Linux, install Python 3.10+ and Ollama. Download and extract RED-Mushaf-Ali-Mir/second-harvest from GitHub. In its folder run `python3 -m venv .venv`, then `source .venv/bin/activate` and `python -m pip install fastapi 'uvicorn[standard]' httpx 'pydantic>=2,<3'`. 2) Keep Ollama running, using `ollama serve` if necessary, and download `ollama pull gemma2:2b`. Start the app with `OLLAMA_MODEL=gemma2:2b python -m uvicorn app:app --port 8001`; open http://127.0.0.1:8001. 3) Describe the meal, the exact parts saved and their condition. Correct the editable food entries before confirming. Supply growing space, sunlight, available containers or potting mix, and your goal. 4) Review the candidate and excluded leftovers. Select an appropriate option, choose Create my growing guide, resolve its missing-information checks, then use Print / Save as PDF to take the guide outside.

How it went

The author demonstrated the planning flow; one guide request took about 79 seconds on a laptop CPU. No growing trial or harvest was reported.

Worth knowing

The model mixed details between food rows. Local weather and planting calendars are not checked; review remains necessary.

Save ideas in this browser. Upvotes count across the board.

Prompt excerpt

No prompt excerpt was provided by the source.

Sponsored

CodeBuddy

Tencent Cloud AI coding assistant for IDE and CLI.

Try CodeBuddy ↗
Explore all →