Fuwn

Applied AI (10,000x) Engineer, forcing gradient descent to larp intelligence

[Fuwn @Fuwned]

I've finally taken the time and was approved for Anthropic's Cyber Verification Program.

Before this, I'd resubmit and play hit-or-miss until Claude would accept my prompt.

[Fuwn @Fuwned]

It's such a strange feeling to see Opus 4.8 seemingly using hardly any of my weekly usage after coming off my Fable 5 spree.

[OpenAI @OpenAI]

We heard you wanted to use Codex rate limit resets on your own time.

Starting today, we’re rolling out the ability to save rate limit resets to use later.

We’re starting Go, Plus, Pro, and Business users with one free reset:

[Fuwn @Fuwned]

Fable 5, btw.

[Fuwn @Fuwned]

Fable 5 is AGI. Me next!

[Fuwn @Fuwned]

Ultracode is incredible. Managing hundreds of agents has never been easier.

[Fuwn @Fuwned]

I'm not sure if I said something to trigger this, but ChatGPT now LOVES using the "😭" emoji. Almost every message contains one.

I don't even use emojis.

[ClaudeDevs @ClaudeDevs]

Claude Code weekly limits are increasing 50%, now through July 13.

Live now for all Pro, Max, Team, and seat-based Enterprise users.

[ClaudeDevs @ClaudeDevs]

Usage limits are up, effective today we're:

1) Doubling Claude Code's 5-hour limits for Pro, Max, Team and seat-based Enterprise plans

2) Removing peak hours limit reduction on Claude Code for Pro and Max plans

3) Substantially raising our API rate limits for Opus models

[Fuwn @Fuwned]

I pulled for the first time on the featured banner.

[Fuwn @Fuwned]

I noticed this some time ago, but Claude Code tends to use subagents much more in its planning mode.

Previously, it would handle everything within its main context, but now it spawns subagents to explore, design, plan, review, and so on.

[Fuwn @Fuwned]

Nice, new colours. Finally, something for my subcortex, not just my cortex.

[Fuwn @Fuwned]

GPT-5.5 no longer appears to spiral out of control when interacting with multiple subagents.

[Fuwn @Fuwned]

@opencode Thanks! I look forward to pushing Kimi K2.6 to its limits.

[Fuwn @Fuwned]

Kimi K2.5 felt cute. I'll have to see how Kimi K2.6 feels after I've finished playing with it.

[Fuwn @Fuwned]

Claude just suddenly died on me about 30 minutes ago. Tried multiple sessions, multiple undos, everything.

Either the service is degraded, or Claude is being difficult about this specific Tailwind CSS problem.

[Fuwn @Fuwned]

Seeing this never grows old.

[Fuwn @Fuwned]

Loving Opus 4.7 (1M) more than anything right now. It's insane at long-horizon tasks, gritty debugging, and its context management has been impressive.

Interested to see how GPT-5.5 will compare.

[Fuwn @Fuwned]

I always forget about Claude Code's recap feature, but I'm glad it exists every time I return to my desk.

[Fuwn @Fuwned]

Does anyone else remember when Claude's first class scripting environment used JavaScript a few years ago?

I always thought that was an interesting choice.