惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Martin Fowler
Martin Fowler
J
Java Code Geeks
博客园 - 【当耐特】
宝玉的分享
宝玉的分享
腾讯CDC
D
DataBreaches.Net
Microsoft Azure Blog
Microsoft Azure Blog
Engineering at Meta
Engineering at Meta
V
V2EX
F
Fortinet All Blogs
MyScale Blog
MyScale Blog
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
T
Tailwind CSS Blog
Jina AI
Jina AI
GbyAI
GbyAI
大猫的无限游戏
大猫的无限游戏
A
About on SuperTechFans
酷 壳 – CoolShell
酷 壳 – CoolShell
爱范儿
爱范儿
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
U
Unit 42
B
Blog
M
MIT News - Artificial intelligence
N
Netflix TechBlog - Medium

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant
The Home Automation Stack That Actually Works (After 3 Ye...
Sam Chen · 2026-05-11 · via DEV Community

Sam Chen

I've been running a smart home for 3 years. Here's what survived the "cool demo" phase and became genuinely useful infrastructure.

The Graveyard of Failed Automations

Before the working stack, let me honor what didn't survive:

  • Motion-activated lights everywhere — turns out, lights turning on when the cat walks by at 3 AM is not "smart"
  • Voice-controlled thermostat — nobody wants to shout at their house when guests are over
  • Automated blinds on sunrise — great in theory, terrible when you're sleeping in on Saturday
  • Smart lock auto-unlock on proximity — security nightmare when your phone GPS drifts

The pattern: automations that are triggered by presence/time fail. Automations that are triggered by intent succeed.

What Actually Works

1. Context-Aware Lighting (Not Motion-Based)

Instead of motion sensors, I use a combination of:

  • Time of day + calendar events (if "meeting" → office lights to video-call preset)
  • Device state (TV on → living room dims to 20%)
  • Manual scene triggers via physical buttons (not voice, not app)

The key insight: automation should eliminate decisions, not create new failure modes.

2. Climate That Learns Behavior

My thermostat schedule isn't time-based — it's event-based:

  • First motion in kitchen (coffee routine) → heat living areas
  • All phones leave geofence → eco mode (this one actually works reliably)
  • Window sensor open > 5 min → pause HVAC for that zone
  • Humidity spike in bathroom → exhaust fan for exactly 12 minutes

3. Security That Doesn't Cry Wolf

After disconnecting 90% of my notification triggers:

  • Only alerts on: door open when "away" mode active, smoke/CO, water leak
  • Zero alerts on: motion (too noisy), camera changes (useless), device offline (temporary)
  • Weekly digest of camera clips rather than real-time notifications

4. The "Leaving Home" Macro

One button press at the door:

  • Lights off (all zones)
  • Thermostat to eco
  • Robot vacuum starts
  • Camera recording activates
  • Door locks (30-second delay for "forgot my keys" scenario)

This single automation saves 10 minutes of "did I turn off...?" anxiety daily.

The Stack

Layer Tool Why
Hub Home Assistant (RPi 4) Local-first, no cloud dependency
Switches Zigbee (IKEA + Sonoff) Mesh reliability, no WiFi congestion
Sensors Aqara (temp, humidity, door) Battery life: 2+ years
Cameras Reolink (local NVR) No subscription, no cloud
Voice None Removed all voice assistants — privacy + reliability
Buttons IKEA Shortcut buttons Physical > voice for reliability

The Non-Obvious Lessons

  1. WiFi smart devices are the wrong choice — Zigbee/Z-Wave mesh doesn't compete with your devices for bandwidth
  2. Cloud-dependent devices will fail — when the company shuts down (it will), your "smart" device becomes a dumb one with no manual override
  3. Your partner must be able to use everything without the app — if automation breaks and only you can fix it, it's a hobby project not infrastructure
  4. Less automation = more reliability — my best setup runs 8 automations, not 80

Building vs. Buying Intelligence

I write about smart home architecture and protocol comparisons at smarthomewizards.com. If you're planning a setup from scratch, the protocol decision (Zigbee vs Z-Wave vs Thread vs WiFi) matters more than any individual device choice.

What's your most reliable automation? Curious what survives long-term for others.