pedromnasc

5 ms·

1.
Frontier models leave a fingerprint, and it's the lab's, not the model's(pedronasc.com)
1 points by pedromnasc 2mo ago|0 comments
2.
by pedromnasc 1y ago
One big problem now is that LLMs are not great at writing Malloy, so it is important to have a intermediate DSL. In the future as the language models evolve or someone creates a fine-tuned model that can write Malloy well, we will be able t
3.
by pedromnasc 1y ago
Exactly. The main thing is that it is easy to underestimate the impact of the context and proper tools. Narrowing down the search space by adding inductive biases into the the system not only make the multi-agent system more correct, but al
4.
by pedromnasc 1y ago
> The "Short Story" section definitely matches my experience at most companies, startups and bigger non-tech companies alike: They already have more data than they're aware of and know what to do with, and understanding wh
5.
Lessons from building an AI data analyst(pedronasc.com)
72 points by pedromnasc 1y ago|12 comments
6.
by pedromnasc 1y ago
Hi all, I wrote a post on some lessons from building an AI data analyst. The gap from a nice demo to a real production system is big -> with a lot of yet to be solved challenges. Would love to share ideas with other builders in the space
7.
Findly (YC S22) Is Hiring a Founding Front End Engineer (London, UK)(ycombinator.com)
1 points by pedromnasc 3y ago
8.
VoxReply – Use Your Voice to Craft Faster Email Replies, by Findly.ai(findly.ai)
4 points by pedromnasc 3y ago|1 comments
9.
by pedromnasc 3y ago
Hey Hacker News! We're super excited to share our latest side project with you - VoxReply. It's an email writing assistant that lets you record your voice and writes the email for you using the powerful GPT-4 language model. Plus