Simplex (@usesimplex) builds developer-first web agents that companies use to integrate with legacy portals. They're already in production, dispatching freight shipments, downloading customers’ invoices, and fetching websites’ internal APIs.
https://t.co/nH0ZccgXpR
Congrats on the launch, @skarpoor12 & @marcon680!
How do transformers do physics? Do they use interpretable human methods or invent “alien physics” that’s hard to decipher?
We find evidence for the former, showing that transformers model the simple harmonic oscillator (SHO) with known numerical methods!🧵@ZimingLiu11@tegmark
At Cursor, we’re fascinated by the problem of deeply understanding codebases.
One useful primitive we’ve been focused on is code graph construction and traversal.
Here's how/why we're tackling this... (1/12)
At Cursor, we've built very high-quality retrieval datasets (for training embeddings/rerankers).
To do this, we use GPT-4 grading and the Trueskill ratings system (a better version of Elo)
Here’s how.. (1/10)
There are some interesting optimizations to consider when running retrieval at scale (in @cursor_ai's case, hundreds of thousands of codebases)
For example, reranking 500K tokens per query
With blob-storage KV-caching and pipelining, it's possible to make this 20x cheaper (1/8)
would love to chat with more @cursor_ai daily users. if you use it daily and hate something about the experience, reply/dm/email/anything and i would love to talk (for ~10 mintues) sometime over the next few days.
Coding just got a little more delightful.
We've raised an 8M seed round, led by the OpenAI Startup Fund to build Cursor!
Read more here:
https://t.co/Sf1Y3RlsVg
@benank_ @Hasnainzelliam @cursor_ai Yep! We have an "AI fix in chat" feature (hover over errors in your code), as well as command + L to give chat the correct error context