#codehumility 검색 결과

Local models can still fool you. The Java code looks clean. The actual payload going to the model is often much bigger than you think. I wrote up how I inspect #Quarkus ↔ Ollama traffic with mitmproxy to see the real JSON, tool overhead, and prompt bloat:


Prove your proficiency and maintain your #100DaysOfCode momentum by earning new learning badges. ✨ For developers starting this challenge, earn the First Learning Pathway and Quiz badge by completing a pathway and deepen your expertise → goo.gle/41piIpV

googledevs's tweet image. Prove your proficiency and maintain your #100DaysOfCode momentum by earning new learning badges. ✨

For developers starting this challenge, earn the First Learning Pathway and Quiz badge by completing a pathway and deepen your expertise → goo.gle/41piIpV

You can learn most high-value tech skills for free online, no degree required. Start with AI (high-leverage skill): fast.ai Learn deep learning practically with real-world applications, not just theory. huggingface.co Access and use powerful

vikas_ai_'s tweet image. You can learn most high-value tech skills for free online, no degree required.

Start with AI (high-leverage skill):

fast.ai  
Learn deep learning practically with real-world applications, not just theory.

huggingface.co  
Access and use powerful

I canceled Claude Code but still have to use it for 2 weeks because it was paid. OMG, it's so dumb! You say, "Why did you do X?" And it's like: "You are right! I shouldn't have done X. I'm now doing Y!" And you're like, "ANSWER THE FUCKING QUESTION!" and it's like, "You are


InCoder-32B-Thinking The first 32B thinking-augmented code model featuring an Industrial Code World Model trained on execution traces across chip design, GPU kernels, and embedded systems. Achieves 81.3% on LiveCodeBench V5 and 84% compile pass rate on CAD-Coder.

HuggingPapers's tweet image. InCoder-32B-Thinking

The first 32B thinking-augmented code model featuring an Industrial Code World Model trained on execution traces across chip design, GPU kernels, and embedded systems.

Achieves 81.3% on LiveCodeBench V5 and 84% compile pass rate on CAD-Coder.

SOMEONE JUST OPEN SOURCED THE MOST COMPLETE CLAUDE CODE SETUP I'VE EVER SEEN 27 agents, 64 skills, and 33 commands. all pre-configured and ready to go planning, code review, fixes, TDD, token optimization. basically everything you'd spend weeks setting up yourself already done

om_patel5's tweet image. SOMEONE JUST OPEN SOURCED THE MOST COMPLETE CLAUDE CODE SETUP I'VE EVER SEEN

27 agents, 64 skills, and 33 commands. all pre-configured and ready to go

planning, code review, fixes, TDD, token optimization. basically everything you'd spend weeks setting up yourself already done

AI transformation requires CEOs to code. I just met a CEO who proved this works. He canceled all meetings. "Meetings are productivity killers." Now he spends every day using Claude Code to automate different departments. Two weekends of coding doubled his team's productivity.


Is anyone still using VS Code instead of switching fully to Claude Code or Codex? Or am I the only one 😅


Few months ago, I set up a small AI hacker team at @Razorpay 2 people. Today, they are 100x builders. With AI, people aren’t the constraint. Org structure is. So now I’m scaling this. If you’ve spent the last few months deep in Claude Code / OpenClaw / agents (or anything


分享一下我的 Agents.md DDIA 和 软件设计的哲学 skill 可以看我之前的帖子

luoling8192's tweet image. 分享一下我的 Agents.md

DDIA 和 软件设计的哲学 skill 可以看我之前的帖子

《pi 的设计艺术》 这本书也可以看了 zhanghandong.github.io/pi-book/ 这本书不是源码导读,也不是 api 文档,我尝试用另一个角度来构建它。 它记录了一个 agent 运行时 ,pi ,在设计过程中做出的关键决策。每一章回答一个设计问题,源码只在需要解释"为什么这样做"时出场。

blackanger's tweet image. 《pi 的设计艺术》 这本书也可以看了
zhanghandong.github.io/pi-book/ 

这本书不是源码导读,也不是 api 文档,我尝试用另一个角度来构建它。

它记录了一个 agent 运行时 ,pi ,在设计过程中做出的关键决策。每一章回答一个设计问题,源码只在需要解释"为什么这样做"时出场。

《Pi 的设计艺术》,也搞上,周末够看了 。。。

blackanger's tweet image. 《Pi 的设计艺术》,也搞上,周末够看了 。。。


✋ stop using commands in Claude Code just use skills. you can get the same DX with: --- disable-model-invocation: true user-invocable: true --- this way you benefit from the entire skill ecosystem and also better cross-compact across CLIs, IDEs. seems like commands are gonna

d4m1n's tweet image. ✋ stop using commands in Claude Code

just use skills.
you can get the same DX with:

---
disable-model-invocation: true
user-invocable: true
---

this way you benefit from the entire skill ecosystem and also better cross-compact across CLIs, IDEs.

seems like commands are gonna

Great resources for learning backend & systems: - Designing Data-Intensive Applications - Beej’s Guides - CMU Database Group - Hacker News - Writing an OS in Rust - ByteByteGo - Computer Systems: A Programmer’s Perspective - Julia Evans’ Blog - Strange Loop conference videos -


📚 Day 265 of learning DSA with @kunalstwt Today's topic: @LeetCode 1 DSA DCC Problem 📌 Key takeaway: Minimum Distance Between Three Equal Elements I 🧠 Feeling: Challenged Here’s what today looked like 👇 #DSAwithKunal #LearnInPublic #100DaysOfCode

ParmanandKS's tweet image. 📚 Day 265 of learning DSA with @kunalstwt 
Today's topic: @LeetCode 1 DSA DCC Problem
📌 Key takeaway: Minimum Distance Between Three Equal Elements I
🧠 Feeling: Challenged
Here’s what today looked like 👇
#DSAwithKunal #LearnInPublic #100DaysOfCode

His ability to explain what’s happening in a way that everyone understands it is 🤯

Judging by my tl there is a growing gap in understanding of AI capability. The first issue I think is around recency and tier of use. I think a lot of people tried the free tier of ChatGPT somewhere last year and allowed it to inform their views on AI a little too much. This is



With nearly 500 GUI applications and a text interface, llambada.com is becoming a new GUI OS where the user says what GUI tool they need and gets this tool instantly. If the tool is missing, an agentic coder builds it for the user based on the text description the

burkov's tweet image. With nearly 500 GUI applications and a text interface, llambada.com is becoming a new GUI OS where the user says what GUI tool they need and gets this tool instantly.

If the tool is missing, an agentic coder builds it for the user based on the text description the

Engineers are so clever. They built claude code, codex, cursor, etc to make their lives easier. No one thought about poor marketers.


🧠 Most valuable lesson 10 years of coding taught me: All code becomes legacy code the moment it's committed. Including the code you wrote 10 minutes ago. #DevWisdom #CodeHumility


Stay humble. Even the best code has room for improvement. 🙇‍♂️💻 #CodeHumility


🙈 That awkward moment when you brag about your "perfectly optimized" code, and a fellow developer points out a more efficient one-liner solution. #CodeHumility #HumbleCoder #OptimizationFail


Devs who don't put effort into reproducing errors or accepting that their code can have errors are the problem! #codehumility


Loading...

Something went wrong.


Something went wrong.