Do the tools in amazon bedrock now replace this approach? E.g. Am I able to achieve this solution in more of a turnkey way using amazon bedrock agents? if so, pros and cons of each approach?
@awssupport17 сағат бұрын
Thank you for your inquiry, Curtis! I encourage following along with this resource, which goes along with this video, for a more in-depth understanding: go.aws/3AB8Z5Q. 🔗 This would also be an excellent opportunity to further expand on your question with our cloud experts on re:Post: go.aws/aws-repost. 📮 They can provide their expertise to address any technical questions that you may have. ☁️ I've also gathered a few additional links that I recommend looking into: go.aws/470c2jS & go.aws/4dA7m6P & go.aws/4fZsCEH. ⬅️ ^TE
@michaelinzo2 ай бұрын
Thank you 🎉
@jeffb63054 ай бұрын
How did you equip the Lex Chatbot with ability to allow user to upload documents? Could the bot participate in Q&A based on the documents the user uploads ?
@awssupport4 ай бұрын
Hi there, glad to see your interest in our video! 👀 I've got a couple resources for you, the first being one of our official docs, check it out: go.aws/4aSEDbD. ✅ Lastly, I'd recommend that you take a look at our re:Post: go.aws/aws-repost. If you like, you can post to and see posts from our industry experts. They have a lot of knowledge and may be able to help you out too. 💡 ^DC
@jeffb63055 ай бұрын
Any suggestions on how to adapt this architecture to include some of the AWS Healthcare tools for a patient facing AI chatbot app?
@awssupport5 ай бұрын
Hello, 👋 these resources are a great starting point: go.aws/48ZqoQZ & go.aws/4cqR2FC. If you have further questions about capabilities related to your use case, you can reach out to our Sales team with this form: go.aws/4coEBde. If the phone number is there, they are available to hop on a call. Hope it's helpful! ^AD
@jeffb63055 ай бұрын
These are great!
@dusanbosnjakovic65888 ай бұрын
How do you make sure to break out of the static flow when necessary? What controls if you will use Claude, llms general answering ability, or static flow, especially once the flow has started?
@awssupport8 ай бұрын
Great questions! These docs may have more info: go.aws/48zY87Z & go.aws/3RXbs0v. Additionally, our community of experts in re:Post might have further insight to your questions, feel free to reach out here: go.aws/aws-repost. 🖇️ ^LG
@dusanbosnjakovic65888 ай бұрын
@@awssupport thanks but that first link wasn't relevant. It was about load balancing.
@awssupport8 ай бұрын
Sorry that wasn't quite right. I suggest reaching out in re:Post for additional advice and tutorials: go.aws/3RWITAo.^RS
@kyleblocksom40557 ай бұрын
Amazon Lex determines user intent based on the prompt (what Lex calls an utterance). Lex intents include statically defined intents and a fallback intent that sends the user prompt to the Bedrock FM if no other intents are identified. The Lambda handler instruments the logic for each intent. For static flows, Lex prompts the user for a series of slot values needed for intent fulfillment. User slot values are validated in the Lambda handler logic. If the user provided slot value is not considered valid, the Lambda handler logic routes the user input to the Bedrock FM then re-prompts for the slot value to continue static intent fulfillment. You can also switch intents by providing a user utterance that maps to another intent based on Amazon Lex NLU.
@jochemtoma9735 ай бұрын
is there a video that goes in detail from scratch to import this solution and make it functional? With a new aws account. Very curious if this is already there or not.
@awssupport5 ай бұрын
Hi there, thanks for the question. ❔ I am going to suggest checking out this: go.aws/4c6CwT6. It should help you get started with generative AI on AWS. Also, a good option is our re:Post. There you can view posts and post to some of our industry experts: go.aws/aws-repost. 🧠 ^DC
@SivaChidambaram156 ай бұрын
So AWS suggesting to use a 3rd party "Kommunicate" to integrate the Lex into the website?
@kyleblocksom40554 ай бұрын
Kommunicate is only used to create the JavaScript plug-in that invokes the Amazon Lex RecognizeUtterance API. Customers will typically integrate Amazon Lex into their existing front ends using custom ReactJS chat components.
@dusanbosnjakovic65888 ай бұрын
You don't mean chain of thought, but ReAct.
@kyleblocksom40557 ай бұрын
"Chain of thought" is intentionally used to described the LangChain conversational agent's prompt reasoning process in relationship to the FM. The action and observation steps are subsequently mentioned in relationship to the agent's available tooling to emphasize RAG capabilities in this example. You can lump these reasoning steps into a single ReAct "chain-of-thought-action-observation" workflow, however, that dilutes the focus on the agent's conversational prompting and action execution relative to LangChain's ReAct agent.