I'm reminded of that adage "Given the opportunity, players will optimise the fun out of a game". A fully automated setup is cool, but if it's working well, there's not much to do. You're not gardening any more.

I have a bunch of plants on my window sill, and considered getting moisture sensors and automating watering, but I realised that kind of defeats the point of looking after the plants myself. On top of the satisfaction of manual work, I recently realised that it's taught me an intuition - the quirks of how much water each plant wants at different times of year (even cacti are thirsty in summer!), and how soil composition and pot size affects water retention. It's probably not generally transferable knowledge, but it was fun to realise I'd come to know my plants really well.

TIL what lakehouse means (a mix of "data lake" and "data warehouse" without the clarifying "data" part). Especially when mentioned in the same phrase as "garden", you tend to think of a physical house first...

I want to do a similar project, but on my father-in-law's house! There I'll have more trouble because the backyard is all covered with cement and the place I want to use is a little distant from the house, so I'm thinking in put some solar panel and battery, there is a roof at this part, to protect the battery and I'll use the roof as a mean to collect water through a gutter. If I put this project I'll write about and share here!

>> I travel quite a bit. Who was going to tell me whether the plants looked healthy? Certainly not my children.

Why not? Why do you keep your family members away from your fantastic project? I would have loved to see something like "my younger kid noticed the leaf ...", "My wife watered the plants ..." etc.

Please remove all tech stuff from your garden. Manage it as a simple human. Let other humans of your family also take part in it.

While your project is technically awesome, you have removed all human factor from it, including yourself. You are letting the AI agents invade the last remaining turfs of the human tribes - the garden work.

You don't need a perfect, highly instrumented garden. That was never the need. Imprecision, lack of full metrics etc are perfectly natural. That's not achievement. That's vulgarity of pouring tons of tech into something natural as a garden, just because you can.

"Lakeshed"

I have just the soil sensor and a smart power plug controlling a plug. Both in Home Assistant and guided by a job in my personal assistant LLM. Much more simplistic. Love your setup!

Too much AI writing; also, the author should have went with some hydroponics to scratch the tech itch.

Great sharing! It's quite fun to read all the story behind the development.

I do love a good goat-skinning project

At least disclose using AI to write the post? The writing is grating to read.

I wonder why Claude didn't suggest him to use Home Assistant from the very start. Pinning sensors in Google Earth sounds - sorry - ridiculous.

That's not lettuce, that's cabbage

why not write this yourself? wouldn't it be more fun? it'd certainly be more fun to read

Spend serious money in several garden sensors for an orchard.

Puts the orchard at the shadow.

Near a Parthenocissus

We all had been there. AI is not real I.

Friends don't let friends to be mocked forever by their partners about the smallest pumpkin violin in the world produced with a Lex-Lutor's level plan. If water is guaranteed by sensors, start by moving this raised bed to a sunny place, buddy. Or plant groceries able to climb.

And for Potato's sake, don't brag in front of your family about your smart stuff before to have obtained a decent result. It always ends the same.

The only good part is that this kale will have bigger leaves at the shadow, but bigger and tastier are inversely proportional in gardening.

Author here. This started as a 2 m² vegetable bed and got out of hand: 8 soil probes, a weather station, a camera, and a Rain Bird valve that I control through an old Android phone tapping its own app.

The part I found interesting: a fixed rules engine for watering kept being wrong in ways that needed context (forecast, last week's response, what the leaves look like). So an agent now runs every morning, reads the full history (4M readings, 62 MB in Iceberg), changes watering within hard limits, and logs what it expects to happen so it can check itself the next day.

Disclosure: the data side runs on LakehouseBox, which I'm building. The code is at github.com/jatorre/garden-lake. Happy to answer anything, including "why not Home Assistant."

[dead]

One practical lesson with agents that control physical things: give them a small set of named actions with hard limits, such as 'water zone 2 for at most 10 minutes', instead of raw device access. Models sometimes misread a sensor value or repeat an action because they forgot they already did it. A log of past actions that the agent reads before every decision, plus a cap on how often each action can run per day, turns most of those mistakes into harmless no-ops. A plant can survive one bad guess but not a stuck valve.

[dead]