wpride

6 ms·

1.
by wpride 3y ago
We do, but mostly as a fairly nice and fairly well-documented interface for plugging together tools like vector DBs and memory with LLMs. My feeling is that with the agentic/AutoGPT hype fading and OpenAI adding the functions API there
2.
by wpride 3y ago
Maybe "complaint" was the wrong word but I disagree with the conclusion that LLMs are "not for trustworthy production systems" for the reasons I stated. Full disclosure, I wrote a blog post called "Text to SQL in Pr
3.
by wpride 3y ago
What is the complaint here exactly? That LLMs aren't enforcing database access best practices for you? That's not their job, that's your job. LLMs generate text. You do the rest. Give the LLM a user with the correct access co
4.
Text to SQL in Production(canvasapp.com)
2 points by wpride 3y ago|2 comments
5.
by wpride 3y ago
Everyone in this space seems to be building on the LSP and classic auto-complete in particular as their UI. But I've found this to be non ideal. - As mentioned in this paper I definitely do not want the AI suggestion crowding out a sug
6.
Using GPT to query your database(canvasapp.com)
1 points by wpride 4y ago|0 comments
7.
Using GPT to Write SQL(canvasapp.com)
2 points by wpride 4y ago|0 comments
8.
by wpride 4y ago
Also, accumulating powerful backers and friends to signal "this is a successful company" instead of building an actual company.
9.
by wpride 4y ago
Phew, was worried when all my CI jobs failed simultaneously
10.
by wpride 5y ago
Not OC but we generate the client types with Apollo codegen ( https://www.apollographql.com/blog/typescript-graphql-code-g... ) using the schema file generated by our NestJS backend
11.
by wpride 7y ago
What news they choose to make News is an opinion.
12.
by wpride 7y ago
We use Contracts too and are in the process of transitioning to Sorbet. In addition to the same runtime type checking as Contracts, Sorbet offers static type checking (and will re-use your runtime signatures in its static analysis).