Last weekend, my brother and I sat spent some time revamping one of our older products, you-tldr (https://you-tldr.com/) to the latest and greatest foundational models. We launched on HN back then, so I figured we'd do it again. The front end is made with Vercel and Tailwind, and the backend is a mixture of GCP Cloud Functions and a Fly.io FastAPI that utilizes LangChain and OpenAI GPT-3.5-turbo API to do the... - Source: Hacker News / over 1 year ago

Did this a couple of years ago: https://you-tldr.com/ Summary remains the elusive hard part...both to do and to serve at scale. - Source: Hacker News / over 1 year ago

Would this give you a similar but quicker result? https://you-tldr.com/ (No affiliation). - Source: Hacker News / over 1 year ago

ML for now is useful for users in terms of images/videos/audios processing, witch can't be done with other means so far, we have some useful example, even if they can't be effective locally (too long training), like https://same.energy things like https://you-tldr.com etc as companion tools to be used when needed, punctually, not at a center of anything... Source: about 3 years ago

I am the founder of Voice In. Voice In lets users type using voice in Salesforce and 1000s of other websites. It leverages state-of-the-art AI models to achieve this. Source: about 2 years ago