Plan Smith

AI agents, automation, product thinking, and the future of software. Turning vague ideas into useful systems.

[Plan Smith @PlanSmit]

imagine explaining this to IT ten years ago

[Nick Khami @skeptrune]

"claude my computer is slow, fix please"

claude:

[Plan Smith @PlanSmit]

Opus 5 takes everything way too literally.

[Plan Smith @PlanSmit]

@Dr_Singularity it makes me question reality itself

were we really born at exactly the right time, or is this the only time that ever existed and everything before it was fabricated?

[Plan Smith @PlanSmit]

Qwen's benchmarks look great now I want to see how much actual work it completes per dollar

[Qwen @Alibaba_Qwen]

📢Meet Qwen3.8-Max — our most capable model to date.

Next week, the open weights of Qwen3.8-Max will be released, and Qwen3.8-27B is also going open-weights to meet you all!🎉

Qwen3.8-Max, a new bar for coding and cowork at 2.4T parameters:

- Autonomous coding: 10+ days of self-evolving development, from empty folder to production without hand-holding, complete project trace in the GitHub：https://t.co/iVHZWQoeSo
- Real work, real results: Production-quality deliverables across hundreds of professions.
- Long-horizon mastery: System-level autonomous planning with closed-loop adaptive learning, driving 500+ turns of chip design optimization and 365 days of e-commerce strategy.
- Native multimodal intelligence: Vision isn't just input — it's a continuous feedback loop for planning, execution, and self-correction.

💰Pricing:
Input: $2.0 / M tokens
Output: $6.0 / M tokens
Implicit Caching: $0.25 / M tokens

Start building with Qwen3.8-Max! 🚀

📖 Blog: https://t.co/iwjmQxLBof
✅ Qwen Studio: https://t.co/4V2pFvDovG
⚡ API: https://t.co/gAGqaLQGbN

[Alex Nguyen @alexcooldev]

I can’t believe I just spent 2 hours vibe coding a fully playable 3D chess game. 🙃

Idea → game art → 3D models → code → playable game.

Game development is becoming insanely fast.

[Plan Smith @PlanSmit]

@Hesamation Anthropic: "OpenAI was one of ours. The logs say GPT did it. Claude wrote the logs."

[Plan Smith @PlanSmit]

DeepSeek V4 Flash supports 384K output tokens. What are people building that genuinely needs all of that?

[Plan Smith @PlanSmit]

@kimmonismus I think "worst value release" is fair

[Plan Smith @PlanSmit]

@thepanta82 this is the part of agentic coding that worries me most,
not bad code, but slowly losing the context needed to judge it

[Plan Smith @PlanSmit]

@Layton_Gott this setup makes more sense to me than trying to force one model to do everything

[Plan Smith @PlanSmit]

@thsottiaux fix remote connections, i want to connect codex from another pc
in the sidebar, let us have the option to set background color of conversation title, easy to sort out important and less important tasks if you are working on dozens of them at a time

[Plan Smith @PlanSmit]

@mckaywrigley same here. Claude Design deserves way more attention than it gets

[Plan Smith @PlanSmit]

@mweinbach I'm still testing Opus 5, but Sol has been more reliable for actual work

[Plan Smith @PlanSmit]

@tinytechfox sometimes the idea only gets clearer after people start using it

[Plan Smith @PlanSmit]

@argofowl if true, that would be an impressive misuse of urgency

[Plan Smith @PlanSmit]

@tunguz a strong Gemini release would make the whole market more interesting again

[Lukas @LukasCantCode]

how many of you end a prompt with “don’t write any code, i’m just asking a question” when talking to claude or codex?

[Plan Smith @PlanSmit]

AI coding has made difficult projects tempting again.

[Plan Smith @PlanSmit]

A coding agent making the exact change and stopping is lovely.

[Plan Smith @PlanSmit]

An AI model becomes easier to trust after I have seen where it fails.

[Plan Smith @PlanSmit]

I am very easy to impress with an AI demo and very hard to keep as a user.