I have tried everything on just about every list of possible treatments, having been in the fortunate position to have made enough money that I can try things full time and being able to bypass doctors for access to medicine and being able to do my own research.
Remember Sonnet 3.5 and 3.7? They were happy to throw abstraction on top of abstraction on top of abstraction. Still a lot of people have “do not over-engineer, do not design for the future” and similar stuff in their CLAUDE.md files.
So I think the system prompt just pushes it way too hard to “simple” direction. At least for some people. I was doing a small change in one of my projects today, and I was quite happy with “keep it stupid and hacky” approach there.
And in the other project I am like “NO! WORK A LOT! DO YOUR BEST! BE HAPPY TO WORK HARD!”
Well my point is that switching to a commit-based workflow with no runtime changes doesn't solve the problem of adobe setup including a malicious commit.
Isolating things to a specific folder is what actually gives any security here, and you can do that on a writable /etc too.
Same experience here. I was working on some easily testable problem and there was a simple task left. In January I was able to create 90% of the project with Claude, now I cannot make it to pass the last 10% that is just a few enums and some match. Codex was able to do it easily.
It hallucinated a GUID for me instead of using the one in the RFC for webscokets. Fun part was that the beginning was the same. Then it hardcoded the unit tests to be green with the wrong GUID.
The hallucinated GUIDs are a class of failure that prompt instructions will never reliably prevent. The fix that worked: regex patterns running on every file the agent produces, before anything executes.
The problem is degradation. It was working much better before. There are many people (some example of a well know person[0]), including my circle of friends and me who were working on projects around the Opus 4.6 rollout time and suddenly our workflows started to degrade like crazy. If I did not have many quality gates between an LLM session and production I would have faced certain data loss and production outages just like some famous company did. The fun part is that the same workflow that was reliably going through the quality gates before suddenly failed with something trivial. I cannot pinpoint what exactly Claude changed but the degradation is there for sure. We are currently evaling alternatives to have an escape hatch (Kimi, Chatgpt, Qwen are so far the best candidates and Nemotron). The only issue with alternatives was (before the Claude leak) how well the agentic coding tool integrates with the model and the tool use, and there are several improvements happening already, like [1]. I am hoping the gap narrows and we can move off permanently. No more hoops, you are right, I should not have attempted to delete the production database moments.
Curious as to how many people are using 4.6, perhaps you’re on a subscription? I use the api and 4.6 (also goes for Sonnet) is unusable since launch because it eats through tokens like it’s actually made that way (to make more money/hit limits faster). I guess it makes sense from a financial perspective but once 4.5 goes away I will have to find another provider if they continue like this :/
Did Zed fix font rendering on low dpi resolution yet? I figured if they couldn't fix such a basic functionality for any editor, for over a year, while pushing AI to please stakeholders, they weren't worth my time. And I question their priorities even if they finally fixed it.
And did they implement debugger support?
When I need a barebones editor I reach for Sublime which doesn't market themselves as something else.
As for Zed taking off, I see a lot of vocals in some niche communities but they barely register, if at all, in large annual surveys.
Numa can do recursive resolution from root nameservers + DNSSEC, .numa local domains with auto HTTPS for dev, and LAN service discovery.
What features would you be interested in?
Split DNS already works — Numa auto-detects Tailscale forwarding rules from the system config. Queries matching .<ts.net> go to Tailscale’s DNS, everything else goes through Numa
If you want to skip Tailscale entirely for home servers, Numa’s LAN discovery auto-finds machines running Numa on the same network. Or add static records in numa.toml for machines that don’t run it.
reply