Analyze Garmin data for meal planning and health patterns
The poster exports Garmin data into a Claude project and meal planner, where Claude spotted a link between eating more on one day and reduced medication efficacy.
Karpathy's Dobby agent controls lights, HVAC, shades, pool and security, alerts on visitors, and replaces six apps.
Karpathy's Dobby agent controls lights, HVAC, shades, pool and security, alerts on visitors, and replaces six apps.
Karpathy built Dobby the Elf Claw as the smart home AI agent. It was given access to control lights, HVAC, shades, pool, and security, with AI used to detect visitors and send alerts. Natural language replaced the six separate apps.
All controls now run through one agent. The source gives no further metrics or shortfalls.
Connect to my smart home devices ([lights, thermostat, shades, security, other]) and let me control them by chat in plain language, like 'dim the living room' or 'set the house to night mode'. Also alert me when [visitor or event] is detected. Test each device with me one at a time, and ask before changing any security or lock settings.
The poster exports Garmin data into a Claude project and meal planner, where Claude spotted a link between eating more on one day and reduced medication efficacy.
A Claude Cowork skill tracks diet and medical metrics in a spreadsheet, and meals told to Claude on mobile are logged in the daily record.
The poster uses Claude with Apple Health, nutrition and lifting app data, plus medication and bloodwork files, and has it maintain an HTML dashboard.