Introducing Hallmark!
An open source design skill to make beautiful UIs and landing pages by default.
Works in Claude Code, Cursor, and Codex.
npx skills add nutlope/hallmark
Stanford just hosted a hackathon. Over 1000 students from around the world came to build for 36 hours straight.
The reward? $100k in prizes.
Here are the winners and crowd standouts we saw at TreeHacks ‘24 @hackwithtrees (🧵):
Massive day for open-source AI.
Plus, developments from Humanoid Robots, Meta, GPT-4V, Google, Stanford Research, and 9 new AI tools.
Here's the rundown of everything going on in AI right now:
Devtools for React Server Components! 🙌
@alvarlagerlof finally took the time to present what he's been cooking!
- Visualize the RSC "wire format" network payloads
- Upcoming browser extension with timeline
💡 Example usecase: find/inspect unnecessarily large payloads
This is huge: Llama-v2 is open source, with a license that authorizes commercial use!
This is going to change the landscape of the LLM market.
Llama-v2 is available on Microsoft Azure and will be available on AWS, Hugging Face and other providers
Pretrained and fine-tuned models are available with 7B, 13B and 70B parameters.
Llama-2 website: ai.meta.com/llama/
Llama-2 paper: ai.meta.com/research/publica…
A number of personalities from industry and academia have endorsed our open source approach: about.fb.com/news/2023/07/ll…
Code Interpreter Beta (rolling out to ChatGPT Plus) is quite powerful. It's your personal data analyst: can read uploaded files, execute code, generate diagrams, statistical analysis, much more. I expect it will take the community some time to fully chart its potential.
To turn on:
In ChatGPT on bottom left click on name > Settings > Beta features > turn on Code Interpreter.
I wrote a bit of a guide to ChatGPT’s Code Interpreter, which I have found to be the most useful and powerful mode of AI.
It is, like every product made by OpenAI so far, terribly named. It is less a tool for coders and more a coder who works for you. oneusefulthing.org/p/what-ai…
The reason code coverage is a flawed testing goal:
100% code coverage is not 100% *code path* coverage
You wouldn’t judge productivity by # of lines of code, why would you judge code quality by % of lines of code?
Wouldn't it be cool if VS code / TypeScript could scaffold out the required properties of a type when creating a value so you don't need to type them all? Does this exist?
This is a super cool resource: Papers With Code now includes 950 ML tasks, 500 evaluation tables (including SOTA results) and 8500 papers with code. Probably the largest collection of NLP tasks I've seen including 140 tasks and 100 datasets.
paperswithcode.com/sota
I’m hoping this will become the basis of regression testing for complicated software involving ML. eg, before rolling out the new version, fuzz searching for discrepancies between new and old version
arxiv.org/pdf/1807.10875.pdf TensorFuzz automates the process of finding inputs that cause some specific testable behavior, like disagreement between float16 and float32 implementations of a neural network