One of the biggest advantages of Tailscale is that your devices stay connected no matter where you are. But sometimes your location should change how your network behaves.
In this video, Jay demonstrates how to use IP source device posture to control access to a subnet router based on the public network your device is connected to. Instead of always routing through your subnet router, you'll learn how to automatically use your local network when you're home and Tailscale only when you're away.
Whether you're running printers, cameras, IoT devices, or other systems that can't run Tailscale themselves, this approach keeps your routing simple while avoiding unnecessary hops.
Documentation
Want to learn more about self hosting, AI, Kubernetes, and modern infrastructure? Come to Tailscale Up and join us in person. Get 20% off IRL tickets with the code ALEXSENTME at https://tailscaleup.com
00:00 Why location sometimes matters
00:29 The problem with always-on subnet routing
00:56 What device posture can do
01:30 Demo: Accessing a printer through a subnet router
02:05 Why routing locally is often better
02:41 Why broader routes aren't a great workaround
03:17 Using your public IP as the condition
03:56 Building the posture policy
05:08 Testing the policy at home and away
05:48 Security considerations
06:15 Final thoughts
From living like a “low-rent Tony Stark” to a few thoughts on Nano vs. Vim, last week’s DevRelCon had no shortage of hot takes. What are your devrel hot takes :face_with_monocle: ?
Tailscale’s DevRel team is normally spread across different time zones, so Alex sat down with Kevin Purdy and Ryan Wuller while everyone was together at DevRelCon in Brooklyn. Ryan explains how a career in video production brought him to Tailscale, while Kevin looks back on more than 15 years of technology journalism, open source rabbit holes, and difficult projects that eventually became stories.
Ryan is beginning his journey into self-hosting with a 3D-printed ThinkCenter NAS, while Kevin wants to explore local AI, privacy, and building useful software for yourself. Tell us in the comments what tutorials, projects, and experiments you want the team to make next.
Get Tailscale:
https://tailscale.com/
Join us at TailscaleUp:
https://tailscaleup.com/
Get 20% off in-person tickets with code ALEXSENTME
There's no shortage of talk about AI. The harder part is what comes after your teams start using it: who's using what, how much of it, and how it's made secure.
Over the next few weeks, we'll be sharing a series of short demos exploring how our centralized AI gateway, Aperture, gives organizations visibility into AI adoption while making it easier to securely connect people to the models and tools they need.
First up: a quick look at what Aperture is, and what it brings to your AI stack.
Learn how to access Home Assistant remotely with Tailscale https://tailscale.com/blog/remotely-access-home-assistant
This step-by-step guide shows you how to install Tailscale on Home Assistant, enable MagicDNS and HTTPS, and securely connect to your smart home from anywhere without opening ports.
We also cover Tailscale Serve, exit nodes, and subnet routes. Tailscale is free for home users and makes secure remote access to Home Assistant simple.
Want to learn more about self hosting, AI, Kubernetes, and modern infrastructure? Come to Tailscale Up and join us in person. Get 20% off IRL tickets with the code ALEXSENTME at https://tailscaleup.com.
In this video, Alex shows how to use the community project DockTail, to turn Docker labels into native Tailscale Services. DockTail watches your Docker containers, reads their labels, and automatically exposes matching containers as private Tailscale Services without publishing ports directly on the host. It is a bit like a Tailscale-native workflow for Docker apps, especially if you are used to tools like Traefik.
Want to learn more about self hosting, AI, Kubernetes, and modern infrastructure? Come to Tailscale Up and join us in person. Get 20% off IRL tickets with the code ALEXSENTME at https://tailscaleup.com.
To clarify, I did, in fact, actually put on an actual suit for this thumbnail. The first time in many years. The sacrifices one makes for their art!
AI agents are everywhere right now but what actually makes something an agent? In this video I break down the agentic AI stack in simple terms, covering models, tools, harnesses, sandboxes, memory, context, and permissions without all the confusing buzzwords.
Towards the end of the video, I also share my self-hosted Hermes and Home Assistant setup, showing how I use Tailscale to securely connect everything together. From fixing Jellyfin over SSH to controlling my smart home with natural language, these real world examples show why agentic workflows are such an exciting step forward. If you want to learn more about AI agents, check out the Tailscale blog post here: https://tailscale.com/blog/agents-are-coming
00:00 - Start
00:48 - What’s an LLM?
05:35 - Aperture
06:24 - Hermes + Home Assistant
Websites—just websites—can still be pretty amazing. I recently rediscovered the joy of a little page with just one job, thanks to Tailscale Funnel. And all it took was reaching a breaking point with my stupid fridge.