๐ฃ Service Update: AWS Billing Console
We are investigating issues with Cost Explorer reflecting inaccurate estimated billing data.
๐ Track status: https://t.co/6qSnkuEzc4
#AWS#AWSSupport
@zachlloydtweets I have seen per user provider limits : default limit is different for different roles , when you hit your limit you need to file a ticket to get an increase.
Used claude code to audit my home network today. It literally browsed my UniFi controller in Chrome, read the zone matrix, and helped me isolate IoT devices from my trusted network. The whole "AI looking at my screen and giving advice" thing is getting pretty useful.
Claude Code planning: "This feature will take 3 days."
Claude Code in practice: one-shots it in 10 minutes.
Me, feeling taunted: that wouldโve taken me a week.
I built a new Python CLI tool called claude-code-transcripts that can create nice readable HTML versions of your Claude Code sessions, both local and pulled from Claude Code for web, and makes it easy to publish them online too https://t.co/pHl8l2lXeK
Just listened to an Apple News podcast narrated by Mark Boyett. A replay of real events about a Hollywood producer, it sound like the beginning of an audiobook caught myself looking for the sci fi angle.
IMO The biggest breakthrough W.R.T coding agents is not the model quality, but the ability to run concurrent agents.
The cost of context switching between issues versus being able to spin up multiple agents with a simple prompt and then massage the results into shape is unmatched.
This probably leans into the โsenior biasโ people have noticed with AI-assisted coding. I already know how to fix these issues, but I can do it faster by kicking off an agent than by context switching myself. I know the architecture, I know how the code should look, and Iโve built up the context to apply good judgment during code reviews.
Now I have a tool that lets me do all of that much faster.
#github #codingagent