Project OnTheSide

[Project OnTheSide @ProjectOnT28479]

@farhanhelmycode No my understanding you are a successful startup owner not a horse 😟

[Tibo @thsottiaux]

if name == "sourabh.gurwani":
return False

And every time a user registers, you start a Codex automation job to add a new if clause (very quickly).

If you want to optimize you can use GPT-5.3-Spark, which is faster.

If you want to optimize further, ask Codex how to implement an index.

[Project OnTheSide @ProjectOnT28479]

@justinpardee @argofowl How is Opus 4.8 with the 100 dollar plan ? I can't seem to find any reviews on the rate limits after the spacex deal, is it better now ?

[Project OnTheSide @ProjectOnT28479]

@shannholmberg Nice now show a demo

[Project OnTheSide @ProjectOnT28479]

@ziwenxu_ @blueemi99 Wait claude code is back to normal usage ? are you working with opus or mainly sonnet?

[Project OnTheSide @ProjectOnT28479]

@ramxcodes Or you can just know what you're doing

[Tom Siwik @tomhacks]

Next time. Can you please just tell us that you're struggling with compute and have to stop subscriptions + do some rate limiting?

While some people will get mad (entitled pricks) most of us would understand and accommodate (albeit a little sour)

Instead this spiel of rug pulling services, free trial pro instead of claude code, lobotomizing the newest release instead of full throttle, banning users and focusing compute on b2b for moneyz.

A little transparency goes a long way with trust and now people found out anyway that Anthropic was struggling (like OpenAI did some time ago too)

You do you, but... c'mon.

[Project OnTheSide @ProjectOnT28479]

@lydiahallie Lydia you're the same person that lied to the whole customer base about issues in compute and blamed ALL of us for misuse and instead of admitting you guys went staright to gaslighting. I hope people learned their lesson and never comback.

[Project OnTheSide @ProjectOnT28479]

@ClaudeDevs So all the gaslighting for about 3 months now, it was all because of compute but you guys never want to admit it you'd rather make us feel dumb or insult us with excuses.

[Project OnTheSide @ProjectOnT28479]

Been facing this error at least 10 times today i am on the 100$ Pro plan this is unacceptable.
@thsottiaux @OpenAI @sama

[Project OnTheSide @ProjectOnT28479]

@MrAhmadAwais @SystematicEdge1 @CommandCodeAI Yeah command code is trash

[Project OnTheSide @ProjectOnT28479]

@bridgemindai Lemme just ignore all the great experiences I keep reading, awesome transparent benchmarks.

And start believing your fuckass platform because you say so.

[SIGKITTEN @SIGKITTEN]

ok what tf codex

[Project OnTheSide @ProjectOnT28479]

@LexnLin

[Project OnTheSide @ProjectOnT28479]

@jasperdevs @sama This is genuinely a skill issue, look up gpt taste skill repo and read it.

[Dwayne @CtrlAltDwayne]

Anthropic: the models are fine

Users: but my token usage

Anthropic: you're using them wrong

Users: my weekly limit is gone

Anthropic: you're imagining things

[Three weeks later]

Anthropic: we found some issues. Here's a one day reset.

[Project OnTheSide @ProjectOnT28479]

@maddada @theo Just a note that it's almost half the token consumption and better on most benchmarks so I'm not defending but seems like a fair tradeoff, it also means it's much faster.

[sui ☄️ @birdabo]

GPT-5.5 now.

[Project OnTheSide @ProjectOnT28479]

@chetaslua Is it 3-4 times faster than gpt 5.4 pro? And will spud have thinking levels as well for speed and intelligence tradeoffs?

[Project OnTheSide @ProjectOnT28479]

@fanjiewang Come on Frank take one for the team.