Anuar Yeraliyev

Hi everyone! CTO of Tyce here. From a technical perspective, our agent has the most deeply integrated tools to a document editor and a file storage. This allows knowledge workers to find information, let AI understand it and synthesize new content and generate it with right format into the editor. And you can still use Tyce with your existing workflow - download from Google Drive, Import and...

Поздравляю с запуском, классный дизайн, интересная концепция продукта

Now every app can add ghibli-fication features to it!

Curious what strategies you are thinking on how to acquire users? As most people will exit the app after finding the match or if there aren't any good options in the area. I also feel like focusing on sublets might be a good idea because they already have posts on Facebook/Craigslist/Kijiji with a location

Wow, this is really big for adoption, it took a while for us to spin up custom editor and implement each editing button in the Toolbar, all the other specs are awesome (table/mobile support, open-source)! Will consider switching in near future for easier maintainability

Instruction following was a huge improvement for us, it can now reliably output html, json better when instructed instead of defaulting to Markdown!

I like how there are different SearchTypes, for insights, information, etc… How do you make the search user-specific? I read on website you personalize to user based on user features but didn’t find it in the docs

Love how open-source models are now beating the closed source ones. Curious if some new use cases will be opened up with 10M context length, previously even with 1M context length it's hard to direct the model what to do and usually accuracy drops.

Love the Developer Build Hours, great to hear from OpenAI devs directly! Even though API docs are really good, there was a gap between all the different ways to use it when building an application, I had to dig into existing open source projects to figure out most common patterns.

Love how simple it is! Was looking for a prompt management and evaluation product. If there was a feature to evaluate prompts/agents/tools on a small dataset, maybe with my custom metrics, that would be killer feature! When we write prompts, we have to run evaluation metrics every time and it's unclear how a single line change would influence UX.