For email, shared infrastructure is key. Because Microsoft needs to see non-spam email from every IP frequently. So your dedicated IP that you use to send 300 emails per month from? You’re landing in the inbox in spite of that, if at all, not because of it. When IPs are shared, it is essential that they not be shared with spammers. No one out there is working as hard as we are to balance shared IP usage, mitigate any reputation issues without any action from our customers, and militantly protecting the IPs against being used to send spam.
All this while we’re not limiting you per domain or per email account, but instead only charging you for the storage allocated to store your email.
We’re not perfect. Our UX leaves much to be desired. But if you’re tired of being a mail admin, and I know you are, there’s only one place where you can hire one that works like you wish you did for pennies.
We use it as our server monitoring system now. Just like “are services online and servers responding” stuff. It’s much more useful than a traditional server monitor as it takes in more context, checks itself for actual network issues to limit false positives, and can be adjusted with just “Hey, do that better next time.”
These days it’s become increasingly important to people that the companies they support have the correct politics. So we’d like to go on record with what MXroute’s politics are:
Fuck the left, right, top, bottom, and everything in between. Bunch of crybabies everywhere.
We deliver email. Whatever Reddit or Facebook post you’re hyperventilating over right now is your problem.
Here is our official response to the Proton controversy:
Screenshots? I don’t trust the word of some person I don’t know when they tell me what happened in a Slack channel that I’m not in.
Of course you pentest AI code. Same as human code. If anyone actually did suggest otherwise, the only correct name to call them would be “moron.” I would be reluctant to say that about someone without actually seeing the conversation.
Anyone not using AI is making a mistake in 2026. At bare minimum you need to have AI doing additional security audits for your code. For decades we said open source was the way to build secure systems because anyone could audit them. Yet here we are in 2026, proving that either people just weren’t doing so or they’re not as talented at finding exploits as AI already is. Your code is being audited by AI, one way or another. It should be you doing it first.
A year ago we called out cold email for it is: spam. Twelve months later, and the sector has attracted approx $100 million in investment, and operators have bought around a million disposable domains to get their mail past filters.
That’s a lot of money to spend on something that’s supposedly legitimate. 🤔
Things couldn't get much worse.
So, we did what we do best: we’ve been tracking that infrastructure and have now listed a large batch of it, with help from a dataset Validity shared with us.
If you’re an ESP, a host or a platform and cold email is going through your network, your AUP should say no and you should enforce it.
Read the full the blog here to find out what happened, and what everyone else in the industry can do about it:
👉️https://t.co/SmE8sWZ39Q
@dawidsowa Bullet dodged on our part. Wish more spammers that like to torture IP reputation and create dramatic statistical correlations with the worst spammers even cared about the policy. Thanks for reading 💜
The internet has always been polluted with mostly stupid bullshit. It hasn’t improved with AI. Now it’s populated with the most well written stupid bullshit you’ve ever seen.
But that stupid bullshit doesn’t come from our network. At least not often. Usually. It happens less than it does elsewhere, put it that way.
We hope this message finds you well.
Gonna keep it closed source. While perfectly within their rights, a competitor appeared with the apparent direct intention of selling at a loss with the idea that they had more money in the bank and could hold out longer than us if they took our market. All while using things we gave them. Lesson learned on being too generous on that side.
@RijnHartman I feel like AI is making people forget that design trends have always been a thing. But maybe everyone is too young to remember the glorious Web 2.0 trend. I’ve always been into design trends, and I especially love the AI ones.
@FBI Apologies. We noticed someone is signing up https://t.co/ATyPxRNhE3 addresses for various websites and newsletters. Only a handful went out from our systems, but we've taken measures to protect against it for now. The measures are mildly excessive and can't remain for too long, but we'll re-evaluate as we go.
AI is the future, and we’re always experimenting with safe methods to implement it. Meet our new Discord support bot. It’s offline now because it created this and no one will ever be able to sleep again.