smivan

My pet theory is that the PRC is heavily subsidizing the hosted inferencing costs in order to capture valuable training data, token usage patterns and the occasional leak of valuable data.

Given their past history and current geopolitics, I'd be willing to bet this is more likely than not.

This is amazing, thank you! Will digest and shoot you a DM with some questions.

I've been trying to figure out how to build up that liquidators network, and sounds like you've really nailed it!

Can you share some vendors and model names? I've been looking into upgrading my home lab from 10g to 100G, but the costs were absolutely astronomical.

I'm likely running a personal lab similar to many folks on HN - about 20-30 wired servers and a small rack of managed Unifi switches.

Appreciate you!

I'm a night owl C-suite (CTO), but with all of my peers as you describe.

It's possible to make it work, you have to be very firm on your boundaries. One option that worked for me was to explain that I can effectively do more timezone coverage despite being in the same Geo-zone.

It also helped to showcase the work I do while they are offline, so when they wake up they see a column of work that has been bounced back across to their plate.

It is definitely hard though, and takes the right crowd who respects you enough to make it work.

It's probably impossible to do this as a CEO though.

Try out various masks. I bought one of each mask available and tried them all, ended up with the N20.

CPAP.com is great for decent prices and direct to consumer sales.

If the internal pressure is annoying, airtight earplugs help with keeping the sinus pressure even.

The challenge that I face is Calibre absolutely craters after 20,000+ ebooks.

I understand that most folks probably have 10-300 books, but in my case I have some large libraries I tried to use with Calibre, and it just absolutely DIES.

The steadfast decision to exclusively use SQLite doesn't help. I opened a PR back in 2016 or so, but the maintainer was quite enthusiastically opposed to changing the DB layer or making it pluggable.

What sort of topologies are you attempting?

I was able to set up fiber link aggregation with triple NICs and a custom VLAN and MTU with no issues. It supports bridges and child interfaces just fine.

Can you share your details?

Congratulations on the launch!

I have been a long time customer of the enterprise version of Streamdal, and I can confidently say Daniel and Ustin are absolutely KILLER engineers. Any time we spoke I always was impressed by their super deep experience and understanding of modern challenges!

So good to see you getting some love on HN. Excited to implement this in some personal projects as well!

Thank you! Will reach out on GitHub with any questions.

This looks amazing! Do you have a sample kubernetes deployment or helm chart we could use? I'd love to deploy the metrics collector on my k3s and have all my servers stream in data.