AutoGen AI Travel Agent TUTORIAL with GitHub Repo!
HTML-код
- Опубликовано: 18 ноя 2023
- This advanced tutorial will explore the Autogen framework for managing multi-agent workflows. We will delve into how Autogen facilitates searching for flights by establishing a connection with the Amadeus API, writing data to a serverless SQL database, and executing queries on this data.
For a deeper dive on this topic, have a look at our blog post that complements this RUclips content: / using-autogen-to-find-...
The GitHub Repo for the code: github.com/john-adeojo/autoge...
Stay updated on AI, Data Science, and Large Language Models by following me on Medium: / johnadeojo
Need to develop some AI? Let's chat: bit.ly/brainqub3-discovery-call
The technical blog on function calling: / how-to-use-openais-fun...
Keen on mastering LLM-app development? Get on our course waiting list: www.data-centric-solutions.co...
Technical blog on function calling: / how-to-use-openais-fun...
Link to previous tutorial on AutoGen: • LLM Projects - A Quick... - Наука
Oh what a lovely thing to see at the top of my feed when I open RUclips. Not even a minute in but thanks in advance!
Hope you enjoyed it!
This is amazing and really easy to follow!!!
Wow ...what a impressive systematic approach to solve a real time problem using Generative AI. This would be a benchmark for many others to solve real-world problem. Thank you so much
Thanks for the feedback. The use case itself isn't quite commercially viable, but I think at some point API calls will be cheaper and the approach could work.
idk man... pretty sure you could do this with a fraction of the memory and a good script... just saying
Very good. Very very good.
Everything in respect to agents is of highest priority to me.
Please provide more content if you can!! I swear to god, I will share, like, subscribe & watch this stuff - just love it.
Thanks, I'll be sure to continue putting out useful content!
I will be dead honest. You are a gem mate. Keep doing what you are doing. Like, shared and subscribed.
Thanks for the feedback!
stopped at 30 seconds to say can't wait. I've been working in python trying to learn simple functions and API calls to a place. I've made a program that will get info from my work's CRM API and then feed AI the response and get things like "how many leads did we have in October." So everything you just said are things I'm learning and struggling with, so I'm excited to see the rest of the video. I've messed with AutoGen some, but I'm ready to start utilizing it.
Cool in depth overview. Great content 🔥
Appreciate it!
pure gold content, thank you
Glad you enjoyed it!
Very nice system one question how do this translate to actually buying or purchasing the tickets based on what the system has found
It's so cool! May I ask a question, autongen function is support other LLM like claude, llama2 ?
It's possible, I'm working on something for Open source LLMs!
Please make tutorials working with free LLMs
This is great!
I do want to ask a question?
Please tell us how this would be better than the usual platforms? e.g. Priceline, Kayak, CheapFlights,etc.
Thanks for the feedback. In this form it wouldn't be better than the established platforms. It's just to show the power of LLM agents to perform complex queries from a natural language prompt.
Very cool! Thanks for the content. Question: Do you have any apps using AutoGen in Production? Things run fine for me in dev, but when I got to Production (Ubuntu/Apache/WSGI), I get all kinds of issues. I'm squashing them, but I'm worried I'll just keep finding new issues running AutoGen in Production.
Great question, I made a video on this! In short, AutoGen isn't production ready. Here's the video if you're interested: ruclips.net/video/YHfN9a8LyYc/видео.html
Great channel my large language brother..
Can you please start putting time stamps? 🤙
Thanks for the feedback. I'll try to incorporate these going forward.
@@Data-Centric the 4 mins of extra workflow will help you build a wider net on here as well. You know sometimes we just want one or 2 functions 🧩 out of the millions. 🤙
I know the topic is Agents, but the results matters. Not sure if this is real data but manually trying google flights it finds WAAAAY cheaper flights and they are shorter ( 17 hours ), for the same dates. $1471 euros for 2 adults, same airports and same dates. Is it because of my location? Because of our searching dates being different? Because we are talking of a difference of almost 50%.
It's using the Amadeus sandpit environment (test not prod data), might be something to do with it.
Where is the api secret ,can only fid the key, you should show the steps of everything n
Also if you use things like sqlalchmey you should tell the suers to install the necessary dependencies etc