An individual could implement most of this today on top of GitHub Sponsors: start a private repo for the book, set it so people who sponsor at a certain tier automatically gain access to that repo, write the book in there as Markdown, accept feedback via issues
CODE
-
MagicPath transforms design images into code in real-time
By
–
One of my favorite things about MagicPath is watching it recreate designs from images.
— Pietro Schirano (@skirano) 23 septembre 2025
The UI streams in real-time, building piece by piece like watching a designer at work.
Image to code in under a minute.
We live in an age of wonder. pic.twitter.com/J1TTuVzaT4One of my favorite things about MagicPath is watching it recreate designs from images. The UI streams in real-time, building piece by piece like watching a designer at work. Image to code in under a minute. We live in an age of wonder.
-
Hyperparameter tuning: logarithmic scaling for better sweep design
By
–
Sometimes I see papers with hyperparameter sweeps over 0.001, 0.003, 0.006, 0.01, etc. Many hyperparameters are better expressed in negative integral log2. Small values like learning rates directly, and values close to 1 like EMA factors and TD lambda / gamma with 1-2**val. It
-
GPT-5-Codex Prompting Guide: Best Practices for Code
By
–
Plus a new prompting guide for the GPT-5-Codex model "Because the model is trained specifically for coding, many best practices you once had to prompt into general purpose models are built in, and over prompting can reduce quality." https://
cookbook.openai.com/examples/gpt-5
-codex_prompting_guide
… -
OpenAI GPT-5-Codex Pricing Details Released
By
–
Here's the OpenAI details page for the gpt-5-codex model https://
platform.openai.com/docs/models/gp
t-5-codex
… It's priced at $1.25/million input and $10/million output – same as regular GPT-5 -
GPT-5-Codex Now Available Via Official OpenAI API
By
–
Looks like it's available via the official OpenAI API now as well – model ID is "gpt-5-codex"
-
LangChain and Manus Partner for Context Engineering Webinar
By
–
Learn Context Engineering Best Practices with LangChain & @manusai We’re excited to share that @RLanceMartin (Founding Engineer, LangChain) and @peakji (Co-founder & Chief Scientist, Manus and recent MIT Innovator Under 35) are teaming up for an upcoming webinar on
-
Production-Grade GenAI Agents: End-to-End Code Tutorials and Blueprints
By
–
@NirDiamantAI
's repository delivers end-to-end, code-first tutorials covering every layer of production-grade GenAI agents Guiding you from spark to scale with proven patterns and reusable blueprints for real-world launches -
Community Fine-tuning Benefits for Model Developers
By
–
I think finetuned models can also be a benefit for the developers. Ie developers may release a base model, and the community applies continued pretraining and finetuning to make a Math, Code, Instruct, Reasoning variant etc so the developers don’t have to.
-
GPT-5 Codex Now Available on ChatLLM and CodeLLM
By
–
GPT-5 Codex Is Coming to ChatLLM and CodeLLM Today