Huge congrats to the team at @imbue_ai! Super excited to see what everyone builds with Scultor! Extremely powerful and intuitive tool, even for those of us that aren't full on engineers.
Sculptor: the missing UI for Claude Code 🎨
Imagine running 5 Claudes in parallel, safely in containers, while you stay in flow. Then bring their work straight into your IDE to test/edit together.
This is how one developer ships like a team. Try it with Sonnet 4.5!
We're finally ready to start talking about what we've been building on the product side over here @imbue_ai. Incredibly fun and talented group of humans to work with. If this resonates with you and want to join us, here you go: https://t.co/AidoWE49a4
Today, AI can generate tons of code—but how do we know if it's good?
That's why we built Sculptor: the first coding agent environment.
Sculptor helps you catch issues, write tests, and improve your code—all while you work in your favorite editor.
Exciting time to be part of the mission here at @imbue_ai. Incredibly talented, driven team training LLMs from scratch on our own cluster. Tons of lessons learned, and sharing some of those lessons with everyone! [shameless plug to apply for jobs here: https://t.co/AidoWE49a4]
Early this year, we trained a 70B model optimized for reasoning and coding. This model roughly matches LLAMA 3 70B despite being trained on 7x less data.
Today, we’re releasing a toolkit to help others do the same, including:
• 11 sanitized and extended NLP reasoning benchmarks including ARC, GSM8K, HellaSwag, and Social IQa
• An original code-focused reasoning benchmark
• A new dataset of 450,000 human judgments about ambiguity in NLP questions
• A hyperparameter optimizer for scaling small experiments to a 70B run
• Infrastructure scripts for bringing a cluster from bare metal to robust high-utilization training
…and more!
Read more and access the toolkit here:
https://t.co/cY4wV6M8pd
Just landed in Heathrow for the UK AI Summit! 🛬🇬🇧
Earlier today, we published a methodology for how government agencies might use large language models to draw insights from citizens' long-form comments.
We also highlight two traps policymakers can fall into. Read more 👇
If you missed our Spacetech Day 2023 livestream, check out the link in our bio for a full replay where you can also see this quick fly-through of the Rocket Production Line in our Skyhawk factory. And stay tuned for an updated factory tour with @NASASpaceflight, coming soon!
📝 We have received our license. @NASA’s ELaNa 41 mission is scheduled for Sat., Feb 5. Launch window opens at 10:00am PT (06:00pm UTC). Stay tuned for updates. https://t.co/Ze8WlLbkB6 #AdAstra
🚀 Live coverage of the 2nd launch attempt for LV0007 begins at T-60 via @NASASpaceflight. The current launch window opens at 9:00pm PT, Nov. 19 (5:00am UTC, Nov. 20).
➡️: https://t.co/Ze8WlLbkB6 #AdAstra
Big week here at @Astra kicking off with the successful test of an industry first fully functional electric propulsion orbital transfer vehicle. Let's go!!! 🚀🚀🚀