This video is part of my Best Selling Udemy Course on Amazon Bedrock and Generative AI (10,000+ Learners) - www.udemy.com/course/amazon-bedrock-aws-generative-ai-beginner-to-advanced/?referralCode=A02153572B8864F928E7&couponCode=NVDPRODIN35
Great videos. Some constructive feedback: it would be useful to see at the end of all this, a webpage where the user inputs their questions and sees the results. Something simple like a search bar and output below. A webpage in React or otherwise. Because everything just in the AWS console seems not practical for an end user of the application. I 've watched a lot of your videos and this is the missing piece.
The lamba function is invoked by looking at the description of yaml file about account id status. How does knowledge base get invoke here to fetch the error description? Does agent automatically invokes everytime both lambda and knowledge base
Great question...When we integrate the Knowledge Base with Bedrock Agents, we are required to provide a description...the LLM of the Agent (Claude-Sonnet) in our case does a semantic match to description provided for KnowledgeBase and OpenSpec API etc. to determine what to invoke...
@@trisalrahul How can we configure AWS Bedrock to ensure that it does not inadvertently reveal cross account information i.e Customer B's information to Customer A, especially in scenarios where Customer A might enter data like in your case account id belonging to Customer B by mistake? Additionally, in a microservices architecture, is it necessary to deploy a separate agent for each OpenAPI integration to enhance data availability ?
@@suchitgupta3110 - Great question...I think its still an evolving space and will really depend on the strategy of the bank, if they want to allow the agent to be accessible post authentication or without authentication or reveal few details pre and some post authentication....You can have multiple services called from same openapi spec and depending on the description, agent can invoke the one where there is a semantic match....
@@suchitgupta3110 Rahul might kept it simple for this tutorial but for real world application there has to more data needed to authenticate the customer e.g. last name + id. More like id + password.
This video is part of my Best Selling Udemy Course on Amazon Bedrock and Generative AI (10,000+ Learners) - www.udemy.com/course/amazon-bedrock-aws-generative-ai-beginner-to-advanced/?referralCode=A02153572B8864F928E7&couponCode=NVDPRODIN35
Great videos. Some constructive feedback: it would be useful to see at the end of all this, a webpage where the user inputs their questions and sees the results. Something simple like a search bar and output below. A webpage in React or otherwise. Because everything just in the AWS console seems not practical for an end user of the application. I 've watched a lot of your videos and this is the missing piece.
Thank you for watching my videos and feedback...Will try and implement your suggestion in future videos....
sir can you teach aws bed rock knowledge base zero to hero
Please check out this video - ruclips.net/video/wN3wmbqLTX8/видео.html
The lamba function is invoked by looking at the description of yaml file about account id status. How does knowledge base get invoke here to fetch the error description? Does agent automatically invokes everytime both lambda and knowledge base
Great question...When we integrate the Knowledge Base with Bedrock Agents, we are required to provide a description...the LLM of the Agent (Claude-Sonnet) in our case does a semantic match to description provided for KnowledgeBase and OpenSpec API etc. to determine what to invoke...
@@trisalrahul How can we configure AWS Bedrock to ensure that it does not inadvertently reveal cross account information i.e Customer B's information to Customer A, especially in scenarios where Customer A might enter data like in your case account id belonging to Customer B by mistake? Additionally, in a microservices architecture, is it necessary to deploy a separate agent for each OpenAPI integration to enhance data availability ?
@@suchitgupta3110 - Great question...I think its still an evolving space and will really depend on the strategy of the bank, if they want to allow the agent to be accessible post authentication or without authentication or reveal few details pre and some post authentication....You can have multiple services called from same openapi spec and depending on the description, agent can invoke the one where there is a semantic match....
@@suchitgupta3110 Rahul might kept it simple for this tutorial but for real world application there has to more data needed to authenticate the customer e.g. last name + id. More like id + password.