Rest in Peace Andy, we love you❤️

Hey Gemini CLI fans! If you're on the free tier and still want access to Gemini 3 we continue to accept submissions to our waitlist: goo.gle/geminicli-wait… Just FYI ... the list is being shutdown with everyone being accepted this Friday! This is your last chance to get in.

It seems a lot of folks hopped on the Claude code train over the holidays. Welcome to the bright side of life! The setup is still a cc terminal with opus 4.5 on one side and another with codex 5.2 high reasoning on the other reviewing each other's output. Welcome

Reached my Claude weekly limits on Sunday. Been forced to use codex for all my needs till my Claude limit resets, and I think it's doing a great job! Love this new phase of engineering. It's like playing video games 😅

😭😭

I'd still rather review every output from an llm. That's the only way I like to use these AI coding agents. Manually. So far, that's worked perfectly for me

Wanna be more effective with AI? Time to embrace a monorepo.

Cracked dev workflow -> Monorepo for shared context + MCP for data and schema context + Plan mode for the detailed implementation plan + back and forth review between opus 4.5 and codex 5.2 high reasoning.

#Java peers…give DT® a follow. He’s a developer based in the UK and is rising up! Please let me know of other hidden figures who i can amplify and make them more seen. “People first. Technology second.”

Joe Muller nah. way too trigger friendly and misses too much stuff unless you do all the charades to keep it in check. gpt is slower but needs much less hand holding.

Garry Tan just ask it to search the web when you need something up to date

Sad to hear about the death of Ron Kenoly. In the 90s he was everyone’s favourite gospel artiste. God bless his soul.

🕊️

There really is a cost barrier to entry when it comes to exploring AI.

Coding agents make TDD so fun. After all, I don't have to write my own tests. I just need to explain my logic to the model. What a time to build

What a goal! Sesko!!

WTF! You really can just build things! #codex

New OpenAI blog explains how OpenAI uses Codex agents plus a tight repo-specific harness of tests, linters, observability, and UI automation to generate and ship large amounts of production code quickly without losing quality. In 5 months, 3 engineers merged about 1,500 pull