ScriptorOfCode

Your Ai planned a 20-phase (tasks) migration plan to merge 2 codebases. After each phase you:

@thsottiaux @sama Make him pay for that. Reset the limits so he suffers!!!!

@sama The hardest the Codex team has ever shipped => fun little bundle. OK Sam. More details here and already available if you update your app! x.com/thsottiaux/sta…

Also, a ton of new Codex features coming soon! Fun little bundle w/the new model.

@confusing_wot who that?

How can I focus if My coach look like that !!!

@SpencerHakimian @grok tldr ple

🚨BREAKING: RETARD POSTED A 470,000 CHARACTER TWEET

MiMo v2.5 + MiMo v2.5 Pro now available in Go • v2.5 → multimodal • v2.5 Pro → built for coding price unchanged

@princesstrixie1 @AesPolitics1 Oh, that's why I was confused. No is winning so, thanks for clarifying that

Republicans fucked around and found out in Virginia lol

@mattpocockuk I don't know (and so does 99%) of developers how to prompt.

What 'advanced' AI coding techniques are you using? I.e. what do you feel like you've discovered that no-one else knows about yet?

@Kappaemme1926 My advice: Get your ass of Twitter and CODE

I’m broke. Literally at $0. I just spent my last $23 on the Codex base plan. I’m a student. I don’t have a job right now. That was literally my last money. So I gave myself a challenge: 30 days to build something that makes money. Not even a lot. Just enough to pay back the subscription. If I can’t… I’ll feel like I failed myself. Day 1 starts now. Let’s see what happens.

@theo Full early report. 1. Gatherer: Great but slow. Successfully gatherer the context needed for my 2 benchmark test. But 2x slower than the current champ Mimo Omni V2. Planner: excellent planner side by side with GLM5.1 and Opus 4.7 Coder: above average, missed a subtask. Auditor

How are people feeling actually using Kimi K2.6 for code?

@opencode Educate me please. Why aren't there more expensive Opencode go plans?

Kimi K2.6 now in OpenCode — Go included

@jullerino @r_marked @t3dotchat Sick!! Let's do opencode go integration and kanban next and I'll marry you

New model picker live in latest nightly! ⌘+Shift+M or /model to open ⌘+1..9 to quick select, or navigate with arrow keys 🔍Search ⭐ Favorite Thanks chrono for the initial work copying @r_marked's design from @t3dotchat

@morganlinton Normal coding: Gpt 5.4-mini-high. GLM5.1 is not a good coder in my experience but a very good planner

Okay, here’s my current AI coding stack. At this point it changes pretty much weekly as the labs are shipping models like crazy now. Planning/specs - Opus 4.7 Normal coding - GLM 5.1 More complex JS/frontend - Opus 4.7 More complex Python/Rust - GPT-5.4 Agentic code reviews - Graphite Right now I’m at a point where over 50% of my coding happens with GLM 5.1, and on Droid. Opus is pretty darn token hungry so I try to use it sparingly. Also finding that GLM 5.1 provides the same output for a lot of things at a fraction of the cost. But Opus is still the goat when it comes to planning and anything you want to look pretty. A month ago I had never used GLM or Droid, things change so quickly!

@Legendaryy Hermes didn't work for me. Stick with Openclaw. Hermes is terrible at orchestration

How’s everyone doing who switched from OpenClaw to Hermes? Did you look back, are you happy with Hermes or are you running both at the same time?

@MilksandMatcha @cerebras If I win I'll call you whatever you want me to call everday for 1 year

Giving away 5 Windsurf Max ($200/month) plans Each person will get 3 months of free Windsurf Max (highest tier). Try out SWE 1.6, Cognition's latest, fastest, and most intelligent model, powered by @cerebras. Winners will be selected from comments in 48 hours, comment below why you want it.

@Amank1412 Because you need a human to blame when it breaks

Backend Developer interview: We can write 80% of our code with AI why should we still hire you. What will be your response?

@vivoplt They need someone they can blame

Why does anthropic have 1,400 engineers if the models are writing most of the code

@rezoundous GLM subscriber for lite program for $36 for a year in Black Friday. Lite plan is now $18 a month.

Bro to bro, don't subscribe to annual AI plans.

@louszbd @ArtificialAnlys ping her

ping me if you're looking to run benchmarks on the glm-5.1 beta

@leetllm @JasonBotterill 200k then 800k of dump zone depending on the model

@JasonBotterill This feels like a classic Bitter Lesson trap. Sonnet 4.6 gives you 1M context on the Max plan, just dumping the whole codebase in usually beats hand-crafted management now. leetllm.com/blog/million-t…

I’ve become a god of fucking MD file context management. I can make Codex comprehend insanely large codebases by having it dynamically write a living record. Honestly it makes 200k of context plenty. You don’t need more if you’re proactive

@Govindtwtt It looks good. What's the problem with that? 90% of non-tech will just see a nice looking website

99% of vibe coded websites look like this