pedromnasc
Joined 7 years ago
Karma
19
Submissions
10
Member since
Jul 26
Recent submissions
Comment · 1 year 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 to have more autonomous agents.
Comment · 1 year 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 also faster.
Comment · 1 year 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 what they have is the starting point before most analysis can be done.
exactly, most of them are concerned about the data they don't have, while in practice they do have a lot to generate good insights.
> Glad I read the post as I hadn't heard of Malloy before. Excuse me if I missed the answer to this, but: How much do you as Findly/Conversion Pattern implement the Semantic Layer on behalf of your users (and if so, I assume you have some process for auto-generating the Malloy models), or do your users have to do something to input the semantics themselves?
We do have an automatic semantic layer generation framework which works as a great starting point, but for the generic case you still have to manually edit / improve it based on the customer's internal context. User's can edit themselves in our UI too, but it usually requires some level of help from us.
We do have a vertical product for commodity trading and shipping: https://www.darlinganalytics.ai/ -> in that case the semantic layer is much more well defined, which makes setup way easier.
Comment · 1 year 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 and willing to learn more about it.
Comment · 3 years 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, you can paste the email you're replying to for better context (optional).
Speak in 50+ languages & choose the tone of the reply With VoxReply, you can record your email in 50+ languages and request that the reply be written in any of these languages. Customize the tone of your response between formal, informal, and business-friendly.
Why we built it:
We were building voice recognition capabilities for Findly.ai (ChatGPT for your Data Warehouse) and thought this was a cool idea.
Email-anxiety relief: We believe it could help people who struggle with email anxiety by providing a simple, intuitive way to draft emails.
Leveling the playing field: Email communication can be challenging for some, especially those with lower levels of education. Voice2Email can assist in creating well-structured emails, making communication more accessible.
Inclusive design: We believe it has the potential to help people with certain disabilities, and we've made the website accessible to those with visual impairments.
We'd love to hear your thoughts and feedback! Try VoxReply now, and let us know what you think!