The program runs six weeks. A small invited group of student ambassadors takes reported bugs and tries to reproduce them, so confirmed issues can be prioritized first. 101 issues reproduced in week one. Every attempt earns a raffle ticket, and a new prize is given each week.
Congrats to Paul Contreras, a Zed student ambassador, who won week one of our Summer Repro Program raffle! 🥳
www.linkedin.com/in/paul-cont... 
Our engineering and design team has been testing 20+ side-by-side comparisons across frontier and open models. All runs are public and open source. Benchmark for this demo here: github.com/CommandCode... We ran a test between the new Qwen3.8-Max, Opus 5 and GPT-5.6 Sol.
3 open models. same prompt. one-shot with the /design command.
Reviewed gameplay, UX/UI and cost.
🔹 Qwen3.8-Max → 9/10 · $0.0480
🔹 Opus 5 → 9/10 · $0.253
🔹 GPT-5.6 Sol → 9/10 · $0.150 
Ran vp dev but meant your dev npm script?
The latest Vite+ now tells you when you ran the built-in command but probably meant your own npm script.
Your future debugging-you & your agents will thank you!
github.com/voidzero-dev... 
Qwen 3.8 Max is now live in Command Code Go.
… and it's going open weight!!
most importantly, next week open-weights of Qwen3.8-Max and Qwen3.8-27B will be released!
🔹 2.4T params MoE flagship
🔹 Input: $2 Output: $6
🔹 Cache: $0.25 / M tokens
3.8 is 20% cheaper than 3.7 
Create custom slash commands and use them frequently.
🔹 Save a prompt as a markdown file
🔹 Define its scope per project or user level
🔹 Run it anytime with dynamic arguments
$ / [arguments] 
Students get a head start: verify your .edu email and you'll get 3,000 credits to start building. This is not just a students challenge, everyone is invited!
Thirty one days. Thirty one apps that teach. Let's build. 🚀
#NetlifyBackToSchool Love our Agent Runners? Use them while completing the challenge and you're in the running for Best Use of AR. Each week the strongest AR build wins 10,000 Netlify credits. Codex, Claude Code, Gemini. Let AI do the heavy lifting.
How it works:
🛠️ Build an app, game, or site that teaches a concept 🚀. Ship it live on Netlify 📥
Submit it so it lands on the showcase wall
🏷️ Tag @netlify.com and use #NetlifyBackToSchool
We'll promote and amplify your builds as you submit them. The mission: build and ship one app per day that teaches one idea. Use what you know best, share your top 3 career tips, or what about a A STEM concept you love?
You don't even need to be a developer to start. Point a coding agent (inside @Netlify.com Agent Runners or otherwise) at it and ship. Summer's over. School is in. 🎒
June was #HotAppSummer. July was #HotARSummer. August is #NetlifyBackToSchool.
Same daily streak, new theme: education. One rule: ship an app that teaches something, every single day.
ntl-back-to-school.netlify.app GPT-5.6 Terra and Luna are 50% off in Command Code.
Get 2X requests for every dollar of credit you spend
Luna:
- $0.10/M input, $0.60/M output tokens
- Available on all plans (Go, Pro, Max, Team)
Terra:
- $1/M input, $6/M output tokens
- Available on Pro, Max, Team
For limited time only. 
PS: To make it easier to try this yourself, I've published a starter kit template here: schickling.notion.site/Personal-CRM... Hopefully this is useful inspiration/guidance for folks who are thinking about building their own CRM. Curious to hear your feedback/thoughts and happy to answer questions!
I've been using my custom CRM for a few months now, and it's been working really well for me. The key unlock is to let agents automatically maintain, so data never goes stale again.
I finally built my own self-driving, personal CRM. After many failed attempts and trying a lot different tools/products, I finally ended up with a system/approach I like built on top of Notion. The best part: It auto-updates, so data never goes stale.
Our engineering and design team has been testing 20+ side-by-side comparisons across frontier and open models. All runs are public and open source. Benchmark for this demo here: github.com/CommandCode... Tested the new DeepSeek V4 Flash, Kimi K3, GLM 5.2.
3 models. 1 prompt. /design command.
Reviewed features, UX and cost.
🔹 Kimi K3: 9.5/10 · $0.0740
🔹 GLM 5.2: 9/10 · $0.0480
🔹 DeepSeek V4 Flash: 7/10 · $0.0005
DSV4 is ~150x cheaper, but rough around edges
Kimi K3 & GLM 5.2 built a polished game 
just learned about the modena pigeon
frizbee is so good!! sponsoring the maintainer now, he's done such good work
Memory updated: user ships on Railway
Changelog #0301
• Railway plugin for ChatGPT/Codex
• dev.new in preview
• 6 new Observability panels
• Fixes and improvements
railway.com/changelog/2026-07-31-railway-plugin-for-chatgpt thank you!