Hi everyone! My name is Will Velida, and I'm a Software Engineer and Cloud Architect!
In my day to day life as a developer, I use .NET and Azure extensively, so join me as I show you how you can build scalable, resilient applications that use Azure, AI, Cloud Native Tech, .NET and many other programming languages!
If you have questions and comments, I'd love to hear from you!
Пікірлер
FTE is missing you, Will. Thanks a lot for a great content !!!
Thank you my friend! This made me smile. Hope you're doing well!
I am trying to run built-in Funplugin, But getting error as "Something went wrong in function invocation. During function invocation: 'FunPlugin-Joke'. Error description: 'No service found.'". How to resolve such errors?
Thanks man, I havent used Github Copilot before but this simplifies work alot.
Great video! Do you know why indexing to Azure AI search uses a different index structure than Elasticsearch or Postgres when using Kernel Memory? Elastic has a indexed “context” field but Azure AI search only has a “payload” field in the index you need to extract the text from?
You got a new subscriber.
Thank you!
Hi Will, I have a question, how to handle the image in local drive?
I want to extract information from a scanned PDF file, how can I achieve
great tutorial .
do you think performance wise that ollama is better than lm studio ? lm studio, while has a nice ui looks like its further away from the metal.
Please, make a video with custom domain and WAF policies.
Subscribed
Thank you!
Clicked on the video to learn about it and realized it's the same thing with FastBert and SQLite. The I integrated it as a long- and short-term memory. Guess I did it the hard way, but I learned a lot. lol. Great video either way!!
what hardware did you use?
Would love to see how this works as well with the Ollama Embeddings API + Semantic Kernel Memory. ❤
Curious to hear if anyone has been able to get local models working with automatic function calling
Same will like to see function calling with local LLMs
Ayyy it's Cecil from Microsoft. Didn't expect you here. What a small world.
@@vivekkaushik9508 Big fan of the channel. Also left Microsoft a while ago 🙂
@@CecilPhillip 😲 Sorry, I didn't know. I hope everything is well.
@@vivekkaushik9508 nothing to be sorry about. It's all good. Still a supporter of a lot of the work going on there
Hi Will. Thanks for the video. Using azure service bus is the same way you did with Azure storage queues? I ve tried to do exactley as you do, provideing a connection string but It didn't work out
I've got a setup with secret retrieval from KV using href, works when KV ingress is public however without having KV open to public it fails to setup complaining its unable to fetch the secret. I've whitelisted the container apps environment IP and allowed it access to KV using user-assigned managed identity. Do you know what the problem could be/what im missing? Thank you.
Interesting way of retrieving registry secrets, any particular reason you are not storing them as environment secrets inside repository settings?
Would definitely love to see more from this series. I have learned a lot, you are a great teacher. Could you do a video about combining vector databases, using embeddings, memory, and RAG with semantic kernel?
0:36
Does anyone know how I can use km in an api? I ran some tests using this great example video but it still doesn't work for me
Awesome video Will. Just one thing, your talking video could be outside main screen to have a full screen view.
Heyy great video here. Will we get more anytime soon ? THNX
As per Microsoft : Handlebars planners will be deprecated in a future release of Semantic Kernel.
you started by creating a .NET 8 project? I can't find that project template available and I have installed the latest version of .net 8 sdk. Is it called something else by chance?
I appreciate Will for this content
Hi, great video! I was tinkering arround and I've found out I was only able to recreate this with Container App with the workload type: Consumption, which doesnt support user defined routes (UDR). When I tried different workload profile It failed with the error "Private Link Service does not support a load balancer that has IP based backend pool, which is used in Consumption + Dedicated environments." I suspect this is caused because load balancer backends uses Ips and not DNS names in different profiles. I was just wondering if you have any tips to tackle this issue. Great content btw, keep it up!
I've beat my head on this one for several days. The instructions in this video will only work with "Consumption Only" and not type "Workload profile". It appears this is a new feature added late last year. The Workload profile/Consumption type allows larger containers than Consumption Only. I believe the only recourse is you need to front the ACE with Azure Application Gateway, then you can front the agw with Front Door.
Could you make a video where you implement a custom domain?
Great series so far! Learned a lot and will continue to check out the other videos in this series. I am looking for a series like this on container apps in a mircorservice setup. Lets say one public facing API URL for service A (p(with either gateway or waf) and 1 or more internal services which A uses to get data. Specificly I am looking for information on how to configure CI/CD in either Github Actions or Azure Devops. I'd like to learn how to implement a canary style deployment pipeline in which a push to main branch deploys on staging, runs load tests and if corrent automatically deploys to canary with 5% traffic, once that has run for a period without errors it deploys to production. Bonus points for basic dashboarding an performance measuring I have not been able to find something like this, so could be an opportunity for you. Please let me know this is something your interested in
Hi there! Great work. I will be starting your series "Lets Build Azure Container Apps!" to learn more. I see you have both the Gateway and WAF as public facing IP. How to decide which one is best for your scenario?
Really useful - thanks Will. Is there a reason you don’t define the resource group in the template, or is it just the setup you had for this walkthrough? Thanks again.
Excellent.
Excellent video. Thaks for sharing Will.
My sincere thanks go out to you for making this playlist and for explaining the concepts so thoroughly.
Great video dude! Just a question, Is planner can be used along side with Semantic Kernel Agent? Say that if I have multiple Agent, and every agent has their own plan?
can we add a web app and an API app as internal containers and access from outside world
Is there an equivalent way to add ImageContent to ChatMessageContentItemCollection in python?
Great video. I expected that the RequestCharge of the x-partition query will be the highest, but if was 0. How so?
Such a great series. Any plans on covering Prompt flow?
It's on the list :) I'll let you know when I publish something. Still deciding whether to break it up into small step-by-step videos, or one long one.
Love these videos. Really helpful in getting up to speed with semantic kernel
Thank you so much! I love what you do for the .NET and Dapr community! Hope you are well!
Thanks for the video! Was struggling to find MusicLibraryPlugin.cs and the other txt. Found them on GitHub in a different repo. But then hit an error at 16:43: "Microsoft.SemanticKernel.KernelException: 'Invalid parameter count for function SuggestSong. 2 were specified but 0 are required.'" So sadly, could not get the demo to work :(
With Semantic Kernel can I build a chat about my own data that is in a database?
Yes! I'm working on some content that covers this which I'm planning to release soon!
@@willvelida super !!! I will be attentive to the new video, thanks
Upvoted! Many thanks, this is really helpful ... Hopefully, access restriction to Azure OpenAI will be solved soon ... 🙂
Thank you! Yeah I'm hopeful that Azure OpenAI will actually be a bit more open soon! 😂
love your videos! keep it up.
Thank you so much!
Sorry to say i could not follow this video
Sorry to hear that. Is there anything you'd like me to explain? I took the Bicep code approach to make it more realistic, but I can set this up via the Portal if that would make it clearer?
You are my bro, you are the man, you are the legend
Thank you! Glad I could help :)
Exquisite presentation skills. Very interesting.
Thank you!
Hey, i found your videos about semantic kernel really helpfull, but i have a problem, i want my gpt35 model to use my azure ai search. the memory thing works and i give the params in the arguments like in the video but it doesnt seem to be able to query my documents. It would be really helpfull if you could make a video about this
nevermind it works lmao
Glad to hear that it works! Let me know if you run into anything else 😊
This whole series is really amazing! Thank you for creating and sharing this content!
Thank you so much! That's really kind of you 😊 If you have any requests on what ACA topics you'd like me to cover, just let me know and I'll add it to the backlog!
@@willvelida My main focus is transitioning from an IT role that deals mostly with on-prem and some Azure IaaS into an Azure specialized engineer or architect role. So any projects that you think would be helpful for that would be awesome!
So handsome
Thank you darling 😁
Thank you so much for these videos! Few suggestions - it would be nice to see what commands you used to pull the client ID, sub ID and tenant ID. I looked up the commands myself but I don't think they're correct because my workflow kept giving an error saying the Client ID I provided was not correct. I couldn't figure out how to fix it so I ended up just creating a new service principal and was able to authenticate that way. Also, the parameters file has an issue, the "containerAppEnvName" doesn't match the "containerAppEnvironmentName" in the main.bicep file. Lastly, I think it would be helpful to show where you look up some of the resources you used so we know where to find them in the future on our own, like the GitHub articles with the template you used for the workflow YML file.
Hey! Thanks for the feedback! I really appreciate it :) I wrote an article which includes the commands that I used in the video. If it still gives you errors, let me know: www.willvelida.com/posts/using-workload-identities-bicep-deployments-powershell/ I'll also double check my parameters file, and I'll add some links that explain the workflow YML file :)