Nicholas C. Zakas

27.6K posts

Nicholas C. Zakas banner
Nicholas C. Zakas

Nicholas C. Zakas

@slicknet

Creator of @geteslint and @bredboxapp. Author. Coach. Real estate investor. GitHub Star. Blog at @humanwhocodes. @[email protected] bsky: @humanwhocodes.com

Boston, MA Katılım Kasım 2008
71 Takip Edilen58.6K Takipçiler
Sabitlenmiş Tweet
Nicholas C. Zakas
Nicholas C. Zakas@slicknet·
I'm excited to share my first new book (or e-book) in over five years! Understanding JavaScript Promises explains not just how promises work but also how to use them in the real world. And the best part? It's free. Grab yours now. bit.ly/promises-ebook
Nicholas C. Zakas tweet media
English
31
288
1.2K
0
Nicholas C. Zakas
Nicholas C. Zakas@slicknet·
I don’t code on weekends. I haven’t in over ten years. Try it. Let your brain relax a bit.
English
12
0
27
1.8K
Nicholas C. Zakas retweetledi
Bredbox
Bredbox@bredboxapp·
This week's Bredbox updates are now live! 📥 Saves is now called Inbox​ to make its purpose clear 🗑️ Deleted links now end up in Trash for 30 days for easy restores ⚡ Fixed a bug that removed at least 2s from each metadata fetch 📱 Test build of the Android app awaiting publish
English
0
1
0
501
Nicholas C. Zakas
Nicholas C. Zakas@slicknet·
Don't sleep on MAI-Code-1-Flash for making small changes. It's incredibly token efficient and now my go-to as I'm making small, iterative changes with Copilot.
English
1
0
3
904
Geoff 🏳️‍🌈
Geoff 🏳️‍🌈@geodbz·
@slicknet I didn’t realize this was a feature added to git and now realize that this is what agents do under the hood when I prompt them on development tasks. Thanks for the write up and your recommendations, this is useful context for what “magic” is happening 💙
English
1
0
0
36
Nicholas C. Zakas
Nicholas C. Zakas@slicknet·
You need to understand git worktrees in 2026. All of the writeups I found confused me. They were either too high level or always focused on AI. So I wrote my own gentle introduction to git worktrees: humanwhocodes.com/blog/2026/07/i…
English
3
4
33
5.5K
Nicholas C. Zakas
Nicholas C. Zakas@slicknet·
GPT-5.6 Luna is blowing me away. What it can get done in fewer than 25 credits is game-changing.
English
2
1
25
13.4K
Nicholas C. Zakas
Nicholas C. Zakas@slicknet·
One of the things I find is that all models are overly defensive when it comes to input validation, often creating a bunch of 'if' statements where they aren't really necessary. This code wants to validate ?permanent=true in a query string. Anything other than "true" should be false. Before/after AI generated code and my edits.
Nicholas C. Zakas tweet mediaNicholas C. Zakas tweet media
English
0
0
7
1.1K
Nicholas C. Zakas
Nicholas C. Zakas@slicknet·
There's a special place in hell for companies that make me spend 10 minutes navigating a phone tree only to tell me to use the website to complete this action.
English
1
0
6
873
Nicholas C. Zakas
Nicholas C. Zakas@slicknet·
Really liking GPT-5.6 Terra. It's much less sycophantic than 5.5 and 5.4. I asked it to make a change that didn't make sense (I had forgot a detail) and it asked if I was sure because of those details. That would have been an annoying mistake otherwise.
English
0
0
5
1.6K
Nicholas C. Zakas
Nicholas C. Zakas@slicknet·
@burkeholland I avoided GPT-5.4-mini and GPT-5.5-mini like the plague. Way too much reasoning for everything. Luna feels like a different class altogether.
English
1
0
0
211
Burke Holland
Burke Holland@burkeholland·
Previously I would have argued that small models are almost always a waste of your time and money. But you can hang out on Luna for a long time before you need to call in bigger guns.
Nicholas C. Zakas@slicknet

So far, GPT-5.6 has impressed me. Luna is the best small model OpenAI has ever produced. Just the right amount of reasoning to accomplish tasks. It's good at SO much. Terra is really good at following instructions explicitly. Both are very efficient in their token usage.

English
5
2
30
4.4K
Nicholas C. Zakas
Nicholas C. Zakas@slicknet·
I evaluate models by credit usage per task: 100, 75, 50, 25, and 10. I try to avoid over 100, and until recently under 10 seemed impossible (MA-Code-1-Flash and GPT 5.6 Luna can do it). When a model delivers a solid PRD or tech spec under 100 credits, I'm impressed.
English
1
0
2
965
Nicholas C. Zakas
Nicholas C. Zakas@slicknet·
So far, GPT-5.6 has impressed me. Luna is the best small model OpenAI has ever produced. Just the right amount of reasoning to accomplish tasks. It's good at SO much. Terra is really good at following instructions explicitly. Both are very efficient in their token usage.
English
1
0
21
6.7K
Nicholas C. Zakas
Nicholas C. Zakas@slicknet·
Argh, Framework 13 Pro is delayed again. Really hoping it's worth the wait.
English
0
0
1
1.8K
Nicholas C. Zakas
Nicholas C. Zakas@slicknet·
If you're like me and were confused by GPT 5.6's naming convention, here's how to remember: - Sol (sun) - largest celestial body in the series - Terra (earth) - mid-size celestial body in the series - Luna (moon) - smallest celestial body in the series I'd still prefer high, medium, and low, but I get the analogy now.
English
3
1
37
12.1K
Nicholas C. Zakas retweetledi
Bredbox
Bredbox@bredboxapp·
We released a bunch of updates this week: 📖 Reading progress is now saved 📦 Import your entire Pocket ZIP​ -- links, collections, highlights 🤖 Automate with n8n ⚡ Faster browser extension saves There's still so much more coming. Join us. bredbox.app
English
0
1
1
467
James Montemagno
James Montemagno@JamesMontemagno·
1 prompt in the #GitHubCopilot app to - Create a full static caption converter website - Create, commit, & push a full repo - Configure github pages, workflow, custom domains 10 mins later. Done captionstack.app
James Montemagno tweet media
English
1
10
47
5.6K

Keşfet

@geodbz @burkeholland @dluzar @elonmusk @BarackObama @taylorswift13 @cristiano @BillGates