rebeccajae

I don't even know if this is an AI thing. I've worked at places that have varying levels of operational maturity, and some of the ones that had very old SRE orgs also had a lot of their SREs operating mostly in the tier 3 given by the article.

One of these orgs had basically all of the noisy pages under control through automated remediation and a culture that pushed towards aggressive automation of pages that had the same root causes. Almost everything I was paged for was some novel or uncommon fault condition. Developing an intuition about how the system worked (well, more specifically, how the system failed) took a lot longer because most things I had to look at were some edge case failure.

I was given a V6 and I bought an adapter that lets me use my dewalt batteries with the vacuum. Kind of a cool party trick but I think they were like $600 new which is bonkers.

I found that Claude would litter the codebase with what felt like notes-to-self. Sometimes it would make an abstraction that wasn't very well thought out and when I pushed back on this choice, it would usually go with something more along the lines of what I suggested in my pushback, but left a comment like "use the git CLI instead of re-implementing git" or something.

It felt like it was commenting on the diff sometimes instead of what the code was doing.

I had ordered something from them and it took me initiating a chargeback for them to ship anything at all. The product that arrived was of terrible quality, and their customer support email is completely useless as well, any email gets a canned "your email is very important to us!" style response. My bank denied my second chargeback request because they had technically shipped something.

It seems technically interesting, but they seem very sparse on details. I don't know if I like the idea of a single unchanging model forever on a chip. How much more expensive would the silicon be if they used rewritable ROM for the weights? Such an arrangement would permit fine-tunes of the model it was designed for, which might minimize concerns about the model becoming outdated.

It sounds neat, but feels like it might fall apart with higher-complexity formats. What does an embedded decoder for a PDF look like? I guess since they are tightly-coupled to the file bytes themselves, the author of the file gets to choose what formats make sense, but not all formats have a one-true-decode-step.

Most people that I have spoken to that are at Google have complained about the internal Gemini agent and seem to believe it's gotten significantly worse lately. Things like it completely forgetting how to call the tools and eventually giving up having wasted a bunch of time, or the agent completely ignoring code style guidance in an AGENTS.md-esque file.

My experience running Gemma 4 locally has been similar: after maybe one or two tool calls it starts making tool calls however it feels like. Just yesterday, I watched it redefine a tool like read_file(start, end) into read_file(start, number_of_bytes) and it refused to even consider that it was wrong.