tim weingarten

This was a fun side project. I used Dispatch in the Claude mobile app while i was traveling last week and built this harness for claude code that takes as input your task, generates a rubric for it, scores Claude’s outputs against the rubric and runs in a loop until the score plateaus. The eval shows a +20pp avg lift over baseline for tasks like writing an investment memo, writing a counter-argument to a claim, designing schema for a billing system and more. It’s available here: github.com/timwein/auto-v…

We asked every version of Claude to make a clone of Claude(dot)ai, including today’s Sonnet 4.5… see what happened in the video

Claude's new ability to work with Excel files is the best I have seen so far I have given it existing spreadsheets to work with and asked it to create new ones. Good use of formatting, formulas, etc. It created all of this, including 406 formulas, from one prompt (& its solid).

Claude-3.7 was tested on Pokémon Red, but what about more real-time games like Super Mario 🍄🌟? We threw AI gaming agents into LIVE Super Mario games and found Claude-3.7 outperformed other models with simple heuristics. 🤯 Claude-3.5 is also strong, but less capable of planning complex maneuvers. Gemini-1.5-pro and GPT-4o perform less well.

A few researchers at Anthropic have, over the past year, had a part-time obsession with a peculiar problem. Can Claude play Pokémon? A thread:

Today, we introduced Claude 3.7 Sonnet and Claude Code! Claude 3.7 Sonnet is our smartest model yet and the first hybrid reasoning model on the market. Claude Code is a command line tool for agentic coding, so developers can hand off complex engineering jobs to Claude. Can't wait for folks to try them out, let me know what you think!

Exciting times at @AnthropicAI — we’re actively hiring for product engineering roles across Growth, Mobile, and more. Open roles here: anthropic.com/jobs let me know if you apply!

Fuyu-Heavy from @AdeptAILabs looks very promising -- this demo shows that the model excels at multimodal reasoning and has a brilliant UI understanding. 🙌Gradio demos are an effective tool to showcase your models' capabilities. Use them to communicate the value of your models.

AI interfaces should go beyond a chatbot: generative UI is next. Here's a demo of a library for LLMs that I've been working on. It prompts models with a vocabulary of rich components described using natural language. Presented at @causalislands last week.

I have literally spent the whole morning on @thehunt. Why is it so addicting! 💁🏼

The @thehunt Android app is in the Google I/O Best-In-Class Design showcase. That's my app. Get on my level. play.google.com/store/apps/col…

@thehunt your the best App in the world 🌎💕

Bringing the heat in the Google Play store with @thehunt Android app. Editor's Choice. Top slot. King of the castle.

lovelovelove this new app i just got called the hunt !! @thehunt

Community-Driven Style panel at #SXSW with @eliza_pb @timweingarten @michelleapersad instagram.com/p/0TAKBpFswo/ #SXStyle

A shopping revolution you'll love! bit.ly/1vS6Q72 @thehunt #TheHunt #Shopping #SocialMedia #Tech #Parents #Dads

The Hunt is an addiction 😍👠👢👗 @thehunt

@thehunt I probably am on your app everyday for at least 20 mins. Am I obsessed??