vineyardlabs

Key driving factors are people who have been laid off working to find positions against the backdrop of many tech companies not adding headcount right now. Additionally the decade+ long messaging campaigns of "CS is going to be the most in-demand major forever etc etc" have now materialized into a steady supply of CS grads every year.

I'm fortunate to be interviewing for a role at a big tech company right now (arguably the most desirable tech employer rn outside of MANGA) and the recruiters have told me that they get almost 1000 applications for every role posted and are basically exclusively hiring from referrals. I imagine this is exacerbated by people using llm scraping bots to automatically submit applications.

With all that said, anecdotally I'm seeing cold approaches from recruiters tick up. 6 months ago my linkedin/email were radio silent, but I've had ~10 people reach out in the last 2 months or so.

Also, it boggles my mind but the alarms/haptics in my S8 seem weirdly unreliable. Like 10% of the time my alarms just decide not to vibrate and I don't wake up. Tons of people have this problem and it seems to have something to do with the cover to mute gesture, but disabling it in settings doesn't fix the issue.

For those two reasons I'm strongly considering ditching my apple watch and grabbing a garmin instead. Just trying to decide if the move outside of the ecosystem is worth it and which model to buy.

If you want more functionality than that, gitlab has a free tier that can be self-hosted. same with gittea.

In 2015 I worked on a piece of software that was related to command and control of the Cassini spacecraft that ran on Sun Solaris machines. That infrastructure would have been first brought up prior to launch in 1997.

Granted, I have to assume that these were not the original physical computers and that there are still ways even today to source new solaris machines for exactly this type of scenario, but still.

Not trying to doubt you, but I find the idea that a company can't terminate an employee for failing to get a clearance for a job that requires a clearance to be tough to believe. You have a source?

The Fed's actions are undeniably painful, but they're also probably necessary to avoid even more pain caused by the last decade of reckless Fed policy and overspending during the Covid recovery.

[1] https://www.youtube.com/watch?v=tCQaMrRuIqA

I/e in a DMA-based ethernet driver, the ethernet MAC receives packets asynchronously from the processor, perhaps faster than the processor can ingest them. So the mac interrupts the processor to give it new packets, and the processor can't sit processing the packets in the interrupt context, so it needs to put them into some ordered list for processing later when it has downtime. In a true embedded system, the memory for this list is going to be fixed or statically allocated, but you still don't really want to have an array-style list with fixed indexing, as you'll have to manage what happens when the index wraps around back to 0 etc, so instead you just construct a linked list in that pre-allocated memory.

I wouldn't say linked lists aren't really used in high-level applications, as I said they're used all over the place whenever you have external asynchronous communication, it's just that modern high-level frameworks/libs totally abstract this away from most people writing high level code.

There's some cursory indication that in the long tail, training LLMs on LLM-generated data causes model collapse. Kind of like how if you photocopy a photocopy too many times the document becomes unreadable.

This isn't really surprising though. Neural networks at large are a form of lossy compression. You can't do lossy compression on artifacts recovered from lossy compression too many times. The losses stack.