All-time opener

Steve here from the OpenAI team–this means a lot! We really hope you enjoy building on it

Hey there! This is Steve here from the OpenAI team–I worked on the Responses API. We have not removed this! It should still work just like before! Here's an example:

https://github.com/openai/openai-python/blob/main/examples/r...

Hi Steve,

I based my assumption on the examples in the documentation[^1]. It is great that we can still use that.

1. https://platform.openai.com/docs/guides/structured-outputs?a...

We've been using this in production for several months, and I'm a huge fan. Unopinionated but incredibly helpful. Big ups to the authors!

<3 from Kysely.

A classic

This is cool - always wondered if something like this was possible/desirable. What do you see as the ideal use cases for Cognate?

Thanks for your comment! I'm currently focusing on making Cognate small and efficient so I can see it potentially competing with micropython as a scripting language for microcontrollers.

We're mainly looking at publicly-traded US securities, including stocks, ETFs, and crucially, mutual funds. Many sources (looking at you, Polygon) don't publish price data for mutual funds. Haven't heard of MarketQA, will definitely check it out

I’ve never really looked at at Mutual Fund data, since personally I use ETFs, and professionally I never had to work with Mutual Funds.

I have no idea if MarketQA has that data. I imagine Bloomberg does, but I couldn’t be sure since I never looked.

Stockevents is an iOS app I saw in HN that from a quick glance at seems to have MF data. I wonder where they source it from.

It is indeed, which is why we stopped after window-shopping Bloomberg and Nasdaq. Seems like it's hard to have cost-effective, high-quality data with and API that's amenable to bulk backfilling.

We have actually looked at Intrinio! Specifically for options data. Again the problem was that the API is not setup for bulk, historical backfills.

No data vendor will give you bulk historical backfills cheaply because then they will be out of business.

I will give you some more names, go pouch a quant or one of their ex-data engineers and maybe you can learn more:

- Bloomberg

- Thomson Reuters

- FactSet

- Refinitive Eikon

There is a reason why so many fintechs are going crypto first. The underlying technology may not be sound but the open business model and accessibility makes innovation a lot easier than dealing with old school financial gatekeepers.

Multi-month, wow. Cool to know this exists though, thanks!

Interesting! I'm no longer a student, so I couldn't get access, but it's cool to know that this exists.

Curious if you've used any of these WRDS data sets? Also really interested to know if you've used the postgres interface to this data, and how you liked working with it as opposed to a regular API-like interface. Thanks!

I have used some of those data sets.

This book demonstrates some usage with R and the WRDS Postgres connector (but obviously you can use anything):

https://tidy-finance.org/accessing-managing-financial-data.h...

PS - some of the data sets referenced in that book are freely available.