Видео

great convo

The technologies for creating AI videos are developing quickly. Discover RunwayML, a feature-rich platform with an intuitive UI and cutting-edge functionality.

Personal expert systems are an interesting area to explore. What if you created five versions of you for a working group, but set different moods or mind sets? If you are a pack rat and have every hard drive you have ever used since the '90s, that is an enormous amount of data to train these virtual me's on.

the system should at least have a plugin or tool layer or block

this was great!

Guy looks like older Messi 😂

poor sound quality

I've been providing tools for RPG Gamers and I came upon the idea of training a SLM to hold information on HOW to use my products... But I don't really understand how this can be done. Ideally, I'd like to have a Discord Bot be able to answer any queries abouts my products, with something like !ask <question>. Behind this Discord bot, an AI / SLM could examine the query and return a response.... That would be awesome! But... HOW the heck do I go about building such?!? I am totally lost LOL

ChatGPT uses a recent type of AI technology that requires huge amounts of computer storage and computer processing power. The hardware system on Microsoft Azure where ChatGPT runs is a very large and complex system of computers and thousands of NVidia GPU processors put together specifically for the needs of ChatGPT. The older AI of the 1950's used computers to try to mimic human intelligence in a very different rule based way. ChatGPT is a brute force mesh capturing pieces of human knowledge in a way that can be matched to prompts to generate answers.

hey david, nice presentation. whats your github for sample codes shown here.

Hello! I'm currently was contacted by your company through Upwork. Do you actually searching for freelancers spanish natives oe it's just scam?

The best part of the video was the graph you almost didnt want him to talk about: ruclips.net/video/sQZLMjaXxoE/видео.html i do this now. i use mistral/mixtra/llama to generate domain specific data to train a domain specific child model. You have to know how to build models and how to code to do this effectively, but the results are pretty awesome!

Bro which microphone use in live podcast

this is a really interesting presentation. But it is so difficult to follow because there is no GitHub or log or any kind of document to go along. I thought I might follow with the code shown on the screen, but there is too much jumping back and forth. Thanks for the effort.

Fucking scammers.. Pay me what you owe me

Fucking scammers PAY ME WHAT YOU OWE ME

These guys are scammers.. They refuse to pay me for a job I did for them and want me to pay to receive my payments.. Fuck hatchworks and everyone involved in this scam

In the eyes of a developer, it feels like a another set of 10 extra hands coding together with you, they might not give you perfectly accurate or even scalable code at the beginning, but it'll definitely do the heavy lifting for you in a way that you can totally expand your productivity rate as well as having more time to focus on making the whole context feeding process more efficient.

You just summed up life💥

I don't think any sane person plans for the end of a conversation.

Might wanna change your company name. People might get confused with the real hatchworks LOL