We just managed to do some black box E2E mobile testing with a #ChatGPT GPT 🤯

Had the pleasure of working with @almouro from @bam_lab on the crazy idea of getting ChatGPT to test an app for us

Took us an afternoon, and guess what ? It worked ✨

More in the 🧵

Just released a Github template to quickly launch a API for #ChatGPT GPT actions powered by @nestframework

✨ Generates the API specs for OpenAI
✨ Local tunnel set up to test in local env
✨ Env vars for easy deployment
✨ Fully documented action example

Link in 🧵

@OpenAI announced they've been working on memory for chats

It was announced 13th of Feb. and I missed it. Why is no-one talking about this ?

Can't wait for it to be available through the Assistant API. It'll simplify so many use cases !

Once again, I must insist that @OpenAI's assistant API is an undervalued tool for prototyping on AI on products

Using my CLI Wrapper, it was able to create this schema of what happens when the user sends a message, just by asking it at the project's root

@GroqInc's speed is wild

Here I'm asking it to read and explain some code
Total inference time is 0.83 s on Mistral 7B

Bigger moment for me than Gemini's 1M token window (although I'm not over the Sora demos yet)

This opens up so many use cases where latency remained an issue

So you can't just fire your customer support team and replace them with #ChatGPT while denying any responsibility ?

Count me as shocked 😮

Just a gentle reminder that @owasp publishes and updates a specific top 10 for #LLM

Make sure to at least read through the short version if you're integrating one in your product

A product manager won't care about a 0.01% increase unless it's about going from 99.99% to 100%

We need to focus the #AI discourse on solving problems, not benchmarks