The new improvements and features coming to @claudeai and #ClaudeCode are awesome but there seems to be so little thought to so many as to how they're to be actually discovered/used.
Would you even know about auto mode if you weren't following Claude Code changelog or tweets?
Claude Code is easy! Just remember:
NO_FLICKER=1
FRAME_SYNC=vsync
USE_BEDROCK=0
MAX_THINKING_TOKENS=64000
REDUCE_MOTION=always
TOOL_COMPOSITOR=metal
claude
--auto
--auto-mode=unattended
--permission-mode acceptEdits
--dangerously-skip-second-thoughts
--quiet-spinner
--strict-mcp
@trq212 "you'll need to explicitly [opt in]"
Please appreciate that when you keep doing this, the majority of users will never know about or experience this functionality.
> Make features opt-in
Maybe having features like "auto mode" not hiding behind ENABLE_FEATURE=1 or --use-feature flags would help?
Claude has so many hidden features that unless you actively keep up (or know someone), it's inevitable you miss some that would be relevant to you
👋 Appreciate the feedback.
Since we introduced Claude Code at Anthropic, engineering velocity has increased hundreds of %, and the rate at which it is increasing is itself accelerating.
The velocity is very much not performative -- we're actively trying to figure out how to build effectively when all of the code is written by Claude. Claude has accelerated the pace at which we ship, and as a result we've been hitting all sorts of new bottlenecks: code review and regression prevention, CI and merge queues, source control reliability, etc. We're working through each of these as they come up, and now have good answers for a number of them.
One of these bottlenecks is figuring out how to best communicate new features to our users. My pov is we need to be doing much better here. The problem isn't that we are releasing quickly, the problem is that we should design features in a way where you don't need to know about them to benefit from them. This is the case for much of what we build, and we need to make it the case for all of it.
To share how we think about it, there's a few ways to approach it from a product design pov:
- Make it so the model can do things for you (eg. enter plan mode, invoke skills, configure your settings)
- Generalize features rather than create new parallel features
- Make features opt-in until we do the above
- Have Claude monitor feature usage and brainstorm/build ways to improve usage while simplifying the system
We try to do all of the above, but as you said, it's not perfect yet, and this is something we're working through. If you prefer a lagging version, you can also use the Claude Code stable release (not latest). We're intentionally being open about what we're seeing, since our customers are seeing the same thing and at least part of our job is helping companies navigate this new way of doing engineering.
Re: source code leak -- it was unintentional, but was also human error. There was a subtle bug that missed several rounds of manual review. We're working on how we can better catch it automatically next time.
@claudeai Desktop still doesn't have Voice Mode, although @claudeai Voice Mode on iOS keeps hearing its own voice and interrupting itself, so not sure I miss it much.
Frustrating comparing how effective I am finding @claudeai Code with my experience with Claude
Desktop
"Chat|Cowork|Code" awkwardly fragments capabilities
Why does it only support a single chat in a single window?
Feels like a different product rather than a cohesive UX
The bug reporting flow in @browsercompany's @arcinternet browser (@joshm) used to be delightful but now involves arguing with an AI chat bot.
Why even have the Early Birds program ?!
I'd just ditch the report if it weren't about the complete break of a primary feature 🙃
RSS feeds powering Live Folders seems like such a natural extension of @arcinternet. New feed items show up as pinned items in the folder, button/automation to remove once read, easy ability to move from live folder to longer-lived folder @ArcMembership@browsercompany@joshm
@arcinternet @kaygalway Definitely makes it more complicated but i think an animated preview (bottom tab shifts over to the other side in sidebar) could make it more intuitive. Would have to try!
@arcinternet Adding a site-specific search feels clunky and seems ripe for improvement. One idea: what if one could right-click a search box (potentially after having a run a search, if technically necessary) to add a site-specific search to Arc?
@Auxbrain any chance we can get better artifact management / bulk consuming? it would take me days of manually consuming one-by-one before the artifacts UI is really usable again
If you're enjoying @standupmaths' YouTube shenanigans I cannot more highly recommend his and pals' Festival of the Spoken Nerd ( https://t.co/l6Ro6BAK7g ). Matt's bonus video with himselves on recursion is worth the entry price alone!
For how long @evernote has been around, I am really surprised at how clunky so many parts of its UX are. For example, want to rename a Notebook you just created? Good luck...