Hashem

I wrangle code by day and craft recipes by night, turning bugs into features and ingredients into feasts! 🍉

[Hashem @HashemKhalifa]

@cursor_ai And the 60$ on cursor what is the equivalent?! Its super confusing…

[Hashem @HashemKhalifa]

@ScriptedAlchemy Same, I let grok bot burn my codex 3 rests overnight using astra high lol.

Still thinking to keep cursor or move to claude only for the usage…

[Hashem @HashemKhalifa]

Lol, but Sam theAI assistance said

I looked at your account again. On a mid-cycle upgrade, leftover credit is applied automatically at checkout. It is the plan price you paid for this period minus the dollar value of included usage already used this period. It is not based on days left in the cycle. Once that included usage reaches the plan price, leftover is $0 and checkout shows the full Ultra price.

[Hashem @HashemKhalifa]

What is happening to OAI in the past few days? Always asking, "May I use this or that? May I deploy, may I, etc"

It was only DOTS doing that, now it's SOL 6.1

It's quite annoying

[Hashem @HashemKhalifa]

@ScriptedAlchemy @GeoffreyHuntley @poteto gtg, getting Cursor Ultra and cancelling my OAI!

[Hashem @HashemKhalifa]

I find it better than Codex to keep astra/sol in lane without drifting, and for purely coding tasks I'm using Devin instead of Codex atm.

[Hashem @HashemKhalifa]

I have been using it with my ChatGPT sub.

It's amazingly good with fusion mode. The main implementer is SOL/Astra, and the sidekick is the unlimited SWE-2

[Cognition @cognition]

You can use any personal ChatGPT plan with Devin

Just connect your Go, Plus, or Pro plan to Devin and all GPT model usage will draw down from you plan's quota

[Hashem @HashemKhalifa]

@ScriptedAlchemy That's amazing, thank you! And much love!

[Hashem @HashemKhalifa]

@ScriptedAlchemy That's amazing, I wish my 62k credits were API usage hahaha
I'm thinking of moving back to Claude... but I've been optimizing my workflow for oai for some time now...

[Hashem @HashemKhalifa]

@realYunfanYe @TejasKumar_ I love it, great work and thanks for building such an amazing project!

[Hashem @HashemKhalifa]

Hi there

[zack @zack_overflow]

You just need that one ADHD person on your team who gets frustrated by every source of slowness and optimizes everything to be faster

[Hashem @HashemKhalifa]

@cheatyyyy I had the same thought, he can't say lol!

[Hashem @HashemKhalifa]

@dabit3 Nader @dabit3, when can I remote work to check my local sessions in the Devin-desktop app?

[Hashem @HashemKhalifa]

@bee_human_ @doodlestein If you want to run commands locally and use ChatGPT to build/deploy and access GitHub without burning your Codex quota, then 100%.Mac's

I can let ChatGPT deploy on my server using my Mac SSH.

It's a bit slower than Codex, but you get unlimited compute!

[Hashem @HashemKhalifa]

@ScriptedAlchemy Lol, it didn’t age well!

[Hashem @HashemKhalifa]

@TejasKumar_ My own automation bots!

[Hashem @HashemKhalifa]

I open-sourced Desktop Relay.

It lets ChatGPT read files and run commands on your Mac, with access controls and activity logs/dashboard.

Connect through Cloudflare Tunnel or your own server.

https://t.co/K454euVrJz

[Hashem @HashemKhalifa]

Working fine so far, now I’ve got two pools of compute: ChatGPT and Codex