61–70 of 81 posts
Earlier quoted context omitted.
The status page is not where you communicate about either the root cause or about the action plan for preventing recurrences.
You sure? That's exactly where I found this. Note the domain :) https://status.cloud.google.com/incidents/ow5i3PPK96RduMcb1S... edit: before some drive-by datamining nerd thinks I do/did SRE for Google, no
Earlier quoted context omitted.
You sure? That's exactly where I found this. Note the domain :) https://status.cloud.google.com/incidents/ow5i3PPK96RduMcb1S... edit: before some drive-by datamining nerd thinks I do/did SRE for Google, no
Fair enough! But that's not real-time communication during an active incident. It's communication O(days) later.
To be fair, too, it's likely been mentioned. I'm biased towards an unreasonable standard due to my line of work.
A status page without some thorough history is glorified 'About Us' :P
Should I be looking at someting like an OpenRouter or AI gateway to ensure uptime for something that currently relies on Anthropic API? Or is there a better alternative to address this availability concern?
Earlier quoted context omitted.
They have. Claude Code was their internal dev tool, and it shows.
And yet even dogfooding their own product heavily, it's still a giant janky pile. The prompt work is solid, the focus on optimizing tools was a good insight, and the model makes a good agent, but the actual claude code software is pretty shameful to be the most viable product of a billion dollar company.
Earlier quoted context omitted.
And yet even dogfooding their own product heavily, it's still a giant janky pile. The prompt work is solid, the focus on optimizing tools was a good insight, and the model makes a good agent, but the actual claude code software is pretty shameful to be the most viable product of a billion dollar company.
What artifact are you evaluating to come to this conclusion? Is the implementation available?
B. Let's just say I didn't write the most robust javascript decompilation/deminification engine in existence solely as an academic exercise :)
Are the vibes off? (pun intended, sorry) I've noticed a trend with their incident reports... "all fixed" , basically. Little mind/words to prevention
The status page is not where you communicate about either the root cause or about the action plan for preventing recurrences.
Every status page incident on every normal company everywhere in the world has links to lead you to the postmortem and their steps to avoid it. Here are a few examples:
https://status.gitlab.com/ -> https://status.gitlab.com/pages/history/5b36dc6502d06804c083...
https://status.hetzner.com/ -> https://status.hetzner.com/incident/2e715748-fddd-427b-a07b-...
https://www.githubstatus.com/ -> https://www.githubstatus.com/incidents/mj067hg9slb4
https://bitbucket.status.atlassian.com/ -> https://bitbucket.status.atlassian.com/incidents/4mcg46242wz...
It's literally a standard for your status page to communicate both about root cause and action plan how to prevent it in the future. Sure, when an incident is just happening, the status page entry doesn't have the postmortem and the steps to avoid, but later on those get added.
Being so overconfidently wrong reminds me of an LLM.
Funny observation - it feels like being in the EU I get a much better AI SaaS experience than folks over in the US. It’s like every other day, the moment US working hours start, AI (in my case I mostly use Anthropic, others may be better) starts dying or at least getting intermittent errors. In EU working hours there’s rarely any outages.
Anthropic has by far been the most unreliable provider I've ever seen. Daily incidents, and this one seems to have taken down all their services. Can't even login to the Console.
I've been paying for the $20/m plan from Anthropic, Google, and OpenAI for the past few months (to evaluate which one I want to keep and to have a backup for outages and overages). Gemini never goes down, OpenAI used to go down once in a while but is much more stable now, and Anthropic almost never goes a full week without throwing an error message or suffering downtime. It's a shame because I generally prefer Claude…
But even when the API is up, all three have quite high API failure rates, such as tool calls not responding with valid JSON, or API calls timing out after five minutes with no response.
Definitely need robust error handling and retries with exponential backoff because maybe one in twenty-five calls fails and then succeeds on retry.