hello sir i need your help on this topic i have elastic beanstalk enviroment i want ec2 instance only ipv6 addresses because aws now charge on ipv4 address but i am stuck on this pls help

Could you please tell me what web server the web application is running on? Is it Nginx or Apache?

Great video! Do you have or know of any guides on how to do this in cdk? Thanks for the great content!

Welcome back! :)

Best video for the topic on the internet. Thank you so much!

Hi txs for your very clear video, i am worry about the ec2 m5.xlarge price for a demo test , any recommendation? Gracias

Hey Trevor, great tutorial, I've learnt a bunch of new stuff and was able to run my Ollama server on GPU, however I've noticed that when I ask purely textual requests Ollama runs 100% on GPU, while if I ask llama3.2-vision to describe an image, it does it 49%CPU/51%GPU, and as such it's really slow. Do you happen to know how we can force llama3.2-vision to use 100% GPU? Thanks!

so helpful. wanna try this. hopefully ec2 free trial can be used to test this

Great explanation and well presented! Thanks for leaving in the parts where you struggle to get it to write something executable. It’s important for us to understand the reality of using these tools and how much actual time will be spent tweaking the prompts. Also gives a realistic sense of the reliability of LLM output.

Could you make another video based on the current state of offerings available for prompt flow?

Thank you so much Sir!! Your tutorials are great!! Newly, thank you!! Subscribed

We can’t local folder be used ??

What an amazing tutorial. Thank you man!!! Super helpful

I am getting an error when trying to use Claude as a model to test the Knowledge Base saying: "Your request rate is too high. Reduce the frequency of requests." Any idea what this might be due to?

This kind of tutorial I wish I had back in a day I learned AWS! As alwasys really informative video!

Guys at 11:47 When it came time to install nvidia-utils, I had problems with the library and driver version (nvidia-utils-550) [message: Failed to initialize NVML: Driver/library version mismatch NVML library version: 550.107]. The solution was to check the package names and I saw that "nvidia-utils-550-server" existed and that's the one I installed and it worked. I assume that the distributions have been updated and now it has to match the driver that was installed (nvidia-driver-550-server).

great tutorial, excellent explanations! thank you sir!

Ran into this guy when i was teaching myself rust a while back, was super happy for him to pop up here as well. Content is super high quality.

Amazon needs to improve their setup instructions. So much time lost until finding your video. Thank you.

Amazing Explanations Thanks!

Awesome. Very thorough and impressive. Thank you.

Great content, thanks! Couple questions: 1. What do you use to track your clipboard history? That was neat 2. Say you generate a flow you really like. How do you productionize it afterwards? Particularly, what is supposed to be the usable artifact of building one of these flows? Is it to build out a library of prompts, and this flow was used just to run the process end to end?

in my rust on aws jorney it turned out that rust sdk indeed requires two hops to fetch credentials, which is not the case for boto3, which worked out of the box in the same container, where i tried to run rust app. now revisiting this video for a deep dive on a topic.

How can I timely invoke agent after a certain, after every 10 min for that too I have to declare the lambda function .

can you share all your install commands and docker compose yaml file?

Informative

Hi sir , please help me.. may i know in aws Console how to check what version if instance metadata service my ec2 is using, i am looking to get information it should show imdsv2.. please let me know where to check in aws Console

Thanks, very clear. I wish it had an option to build and deploy from local machine.

Nice walkthough; question: is it a requirement to use IPV6? Can we use v4 instead?

Can’t we change the output of the prompts so that we can supply that output to the iterator? Currently it’s not allowing it to pipe with iterator as iterator is expecting array as the input type

Very great tutorial, was looking for the exact same thing (y)

Can't we use the hosted OpenWebUI web interface to download additional or remove existing model instead of exec'ing into the docker container ?

Why not use Inf1.xlarge for the EC2 instance type instead?

How much cost for this external private CA bundle?

What’s is the cost for acceptable spec.

This was really helpful. Very clear. No rushing through. Can you make a video on how to register both on-prem database like Postgres and RDS database to AWS Datazone?

Spot on. Thanks! 🙂

Brilliant!

Don't this have any memory concept suppose i say my name is shubham and in next query i ask it what's my name

Nice! Thanks for your good initiative.

I think you could set "OLLAMA_BASE_URLS=ollama:11434" for the WebUI (instead of using the IP)

Your conversation flow is smooth as Prompt flow :). Good job mate. Subscribed

the previous video for me was like clicky-clicky in the console, but this one makes definitely more sense since due to integration with lambda (as you mentioned) the use cases are limited by imagination. I am not yet that familiar with bedrock api and i ask myself how can i integrate this feature in a real project, since neither me, nor the users of my app will login to AWS to play with the workflow. Would be great if you could also mention this in future videos. Thanks for great content! Keep going!

Interesting one. I am curious how complicated the flow will be in real life scenarios and if it’s manageable with all those prompt and debug units

Appreciate for producing this content.

Great video Trevor! If you ever do a follow-up on how to use a KB with an Agent that'd be great. FYI, I'm working on a recommendation engine that gets data from a Bedrock KB and wanting to have it as a frontpage bot. For newbies, it is a bit confusing how to set this up. A prompt flow allowed me to clean the kb results via a lambda but a prompt flow doesn't give me the ability to subsequently drill down on the resultset: it has no memory. An agent has memory but how do I guide it along with the need to clean up the results from a kb via lambda etc. I'm going in circle.... Anyway, I love your videos, thank you for posting them! 😊

Nice explanation

Great video. Afaik the eventbridge does guarantee event order. I need to create a simple architecture where we have a HR system (not known yet) where I would like to use a event based architecture. But the order of the event must be correct, like employee created -> employee updated -> employee deleted. I then need to have several system receiving these events.

Good clear content. Was really helpful for me. Thank you.

Thanks for the nice demo. If there is no matched rule, where does the incoming message goes in the event bus? Is it stored somewhere for sometime?