Oh this is timed perfectly - I was just writing my own erlang agent framework but this is much better!

As a mobile dev at YouTube I'd periodically scroll through crash reports associated with code I owned and the long tail/non-clustered stuff usually just made absolutely no sense and I always assumed at least some of it was random bit flips, dodgy hardware, etc.

I have max, I kept hitting usage limits. I have 4 projects that I will code in parallel, so while one agent is working, I spin up other agents to complete things. Most of my effort now is designing agile roadmaps with specifications, epics, sprints and implementation cards (using AI to create it, then reviewing it), so the Agents have a massive, detailed roadmap. I review code but I also built a framework where much of the code is generated by templates not the model itself so the review is mostly cursory.

Many of us have both Max and use extra-usage

I assume you're doing things with the API that aren't coding tasks that could be done with Claude Code? Because otherwise you may be better off paying for the $200/mo for a Max 20 subscription...

2. fun

ZFS snapshots can be used to similar effect, basically for free.

Or btrfs for that matter. I'm doing something similar with btrfs. Used zfs for a while, but the external repositories kept getting out of sync with the distribtion kernel, so system updates required manual intervention. That annoyed the heck out of me over time. Switched back to btrfs, which has been working fine for the last year. 10 or so years earlier I still had data corruption and bugs with btrfs.

Kagi assistant includes all Anthropic, Google, OpenAI and Grok models as well as all the common open weights models.

The act doesn't explicitly require watermarking, does it?

Even better is the live version: https://vimeo.com/714215610

Thanks! That was actually really good. His performance matches the emotions underlying the song.

Something all the covers of Behind Blue Eyes by The Who seem to miss is the slowly rising anger and frustration.

Sonnet 4.5 was a pretty significant improvement over Opus 4.

Yes but it’s easier to understand difference between 4.5 sonnet and opus and apply that difference to opus 4.6