JavaScript Indonesia Channel

Every AI code editor wants to replace VS Code, but the smartest ones are just extending it https://www.xda-developers.com/ai-code-editor-replace-vs-code-smartest-extending/

Claude Is Getting Dumber And Open Models Are Getting Smarter — So Why Haven’t You Switched Yet https://andrewzuo.com/claude-is-getting-dumber-and-open-models-are-getting-smarter-so-why-havent-you-switched-yet-f0bef0eddb80 —- Anthropic’s Claude Opus 4.7: What Actually Changed in 48 Hours https://blog.stackademic.com/anthropics-claude-opus-4-7-what-actually-changed-in-48-hours-101dfed8fde2 —- GPT-5.5 vs Claude Opus 4.7: Which One Should Rails Developers Actually Use? https://levelup.gitconnected.com/gpt-5-5-vs-claude-opus-4-7-which-one-should-rails-developers-actually-use-889bbf2201cd

Claude Code Best Practices https://uxplanet.org/claude-code-best-practices-57fbc3d5d09e —- AI Didn’t Kill Engineering. It Made Bad Engineers Visible. https://medium.com/vibe-coding/ai-didnt-kill-engineering-it-made-bad-engineers-visible-3bd53a7b3a2c —-

DESIGN.md: The Complete Guide https://medium.com/autocomplete-real-world-ai/design-md-the-complete-guide-ac8c5cd0d14d —-

Baru nemu project keren di GitHub Gemma Chat, app buat "vibe coding" yang jalan 100% offline di Mac Apple Silicon. Jadi gini konsepnya: kita ngomong aja ke AI apa yang mau dibikin, terus AI-nya generate kode (HTML/CSS/JS) langsung dengan live preview. Dan yang bikin menarik, semuanya jalan di lokal, ga perlu internet sama sekali setelah model pertama kali didownload. Pake model Google Gemma 4 via Apple MLX framework. Ada 4 pilihan model dari yang ringan (~1.5GB) sampai yang berat (~18GB). Fitur favorit: • Fully offline, cocok buat coding tanpa internet • Live preview real-time saat AI nulis kode • Speech-to-text pake Whisper (WASM) • Setiap percakapan punya workspace sendiri Tech stack-nya: Electron + React 19 + TypeScript + Tailwind Yang lagi cari cara coding tanpa tergantung API key atau cloud, worth buat dicoba. Open source juga, MIT license GitHub: https://github.com/ammaarreshi/gemma-chat

If AI Can Code,Why Are Big‑Tech Founders Coding Again? https://ai.plainenglish.io/if-ai-can-code-why-are-big-tech-founders-coding-again-6791526cde51 —- Forget ChatGPT & Gemini — These New AI Tools Actually Solve Real Problems (No BS) https://medium.com/@nitinfab/forget-chatgpt-gemini-these-new-ai-tools-actually-solve-real-problems-no-bs-1c5d1ca6a185 —- Claude Opus 4.7 vs OpenAI Codex: Same Day, Opposite Bets https://medium.com/@pankaj_pandey/claude-opus-4-7-vs-openai-codex-same-day-opposite-bets-3a8422510064

I Spent 3 Days Researching Self-Hosted AI. Here’s Why You Should (And Shouldn’t) Actually Do It https://medium.com/ai-threads/i-spent-3-days-researching-self-hosted-ai-heres-why-you-should-and-shouldn-t-actually-do-it-e62ad7fcd9f1 —- Tested Kimi K2.6 vs GLM-5.1 on 15 Real Coding Tasks — The 0.2-Point SWE-Bench Gap Hides a 43% Price Gap (and the "Winner" Codes 11 Points Better) https://pub.towardsai.net/i-tested-kimi-k2-6-2daa40001fd6

AI now writes 97% of my code. Here's what I learned https://swizec.com/blog/ai-now-writes-97-of-my-code-heres-what-i-learned/?ref=dailydev —- Software engineering may no longer be a lifetime career https://www.seangoedecke.com/software-engineering-may-no-longer-be-a-lifetime-career/?ref=dailydev —- Before You Build Anything Hard With Claude Code or Cursor, Do This First https://medium.com/realworld-ai-use-cases/before-you-build-anything-hard-with-claude-code-or-cursor-do-this-first-85156ae9a0ed

15 Essential Sections Every README Needs: Give Your Project What It Deserves https://dev.to/georgekobaidze/15-essential-sections-every-readme-needs-give-your-project-what-it-deserves-fie?context=digest

Git Workflows That Won’t Make You Cry: A Dev’s Cheat Sheet https://dev.to/ajay_mudettula/git-workflows-that-wont-make-you-cry-a-devs-cheat-sheet-3663

The Fastest Way To Become an AI Engineer (For Software Engineers) https://medium.com/data-science-collective/the-fastest-way-to-become-an-ai-engineer-for-software-engineers-b5ec83c11e25

20 MCP Servers That Turn Claude Into a Senior Engineer https://medium.com/tech-and-ai-guild/20-mcp-servers-that-turn-claude-into-a-senior-engineer-db985e414b06 —- I Tested Claude Opus 4.7 vs 4.6 on 7 Real Tasks: The Default Setting Swap That Quietly Ends the "Nerfed Claude" Era https://pub.towardsai.net/i-tested-claude-opus-4-7-vs-4-6-on-7-real-tasks-the-default-setting-swap-that-quietly-ends-the-6757e3c18418 —- Microsoft Quietly Killed Opus on the $10 Copilot Pro — Here's the Math on Whether You Should Cancel https://pub.towardsai.net/microsoft-quietly-killed-opus-on-the-10-copilot-pro-heres-the-math-on-whether-you-should-cancel-61af8f4fa76b —— I Tested Kimi K2.6 vs GLM-5.1 on 15 Real Coding Tasks — The 0.2-Point SWE-Bench Gap Hides a 43% Price Gap (and the "Winner" Codes 11 Points Better) https://pub.towardsai.net/i-tested-kimi-k2-6-2daa40001fd6 —- GPT 5.5 vs Opus 4.7: Which AI Model Is Best for Product Design? https://uxplanet.org/gpt-5-5-vs-opus-4-7-which-ai-model-is-best-for-product-design-23752483196a —- GPT 5.5 vs Opus 4.7: Which AI Model Is Best for Product Design? https://medium.com/no-time/gpt-5-5-vs-claude-opus-4-7-which-one-should-you-actually-use-7812d916155c —- Anthropic Promised Claude Opus 4.7 Would Change Everything. Here’s What Actually Happened. https://pub.towardsai.net/anthropic-promised-claude-opus-4-7-would-change-everything-heres-what-actually-happened-bc3bc18fafa6

New Claude Code Desktop App https://uxplanet.org/new-claude-code-desktop-app-77d198553e9a —- Step-by-Step: Complete Auto Mode configuration in Claude Code https://medium.com/@dan.avila7/step-by-step-complete-auto-mode-configuration-in-claude-code-2ca3a0267a08 —- Spec Driven Development — Three Maturity Levels Every AI Team Should Know https://medium.com/@wasowski.jarek/spec-driven-development-three-maturity-levels-every-ai-team-should-know-648c93cf1e1d —- Claude Design Beats Every AI Design Tool! https://medium.com/design-bootcamp/claude-design-beats-every-ai-design-tool-c8d43275bebd —-

The New Definition of Software Engineering in the Age of AI https://www.freecodecamp.org/news/the-new-definition-of-software-engineering-in-the-age-of-ai/

How AI Changed the Economics of Writing Clean Code https://www.freecodecamp.org/news/how-ai-changed-the-economics-of-writing-clean-code/

DeepSeek-V4-Pro is a frontier Mixture-of-Experts model with a 1M-token context window and three reasoning modes. ollama launch opencode --model deepseek-v4-pro:cloud https://ollama.com/library/deepseek-v4-pro

8 JavaScript Practices That Make You a More Reliable Developer https://javascript.plainenglish.io/8-javascript-practices-that-make-you-a-more-reliable-developer-8e729e4aba70