🌟 Friday Virtual Coffee: How to build a chat bot API with LLM function calling and AWS Lambda 🌟 We’re excited to invite you to our upcoming webinar this Friday, from 1:00-2:00 PM PDT, where we’ll be diving into a fascinating topic: "Fleak LLM Function Call with AWS Lambda." Sign up HERE: https://lu.ma/e1ma4qi4 💖 Why Attend? We all know how to "prompt" an LLM to do stuff by now, thanks GPT, thanks Claude. But with all the "agentic" workflow talks on the street, have you ever wondered what is that all about? This webinar is going to demystify how to create a simple AI agent for you by simply switching on the "function call" mode in LLMs. You will be surprised that you are already fully capable of doing all of that with your existing engineering knowledge. 💯 Why it can be useful for you? Imagine walking into a critical team meeting on a Monday morning and needing a quick summary of last week's discussions. Instead of sifting through hours of team chat manually, what if you could retrieve and analyze the most relevant points in seconds? At Fleak, we faced this challenge head-on and developed an internal workflow API using AWS Lambda and Large Language Models (LLMs) to do just that. #API #functioncall #llm #openai #awslambda #serverless
Fleak’s Post
More Relevant Posts
-
Cloud Security Chick | LetsDoDevOps.com | Microsoft MVP (DevOps) | AWS Community Builder x2 (Containers) | Day Two DevOps Podcast Host | Former Hashi Ambassador | Open Source Zealot
I got a) very excited and b) very caffeinated this weekend and updated my Slack AI Bot to do a bunch of new cool things. She (I lovingly call her Vera) relies on Claude v3.5 Sonnet v2 on AWS Bedrock for smarts). She can now: - Support single or multiple files shared from slack through to the model (fetching and base64 encoding on the way) - Recognize and respond to other file types that aren't yet supported. x Most image types are supported x PDFs are coming! Bedrock Claude doesn't support yet - Recognize and understand multi-person thread conversations x I shim in the "real name" (from slack) of each person speaking into each message sent to the model, here's the relevant code. o Needed an additional slack permission for the bot to find user data from the User ID that is passed in message events x This permits her to understand that there are different speakers, who have different opinions, and she can address individual viewpoints and identify overlaps. Big hopes here. - As an unintended (but welcome!) consequence, Vera now knows your name when you speak to her, she'll greet you by name. Here's an example of me asking her to describe an image, her recognizing that it's an architectural logical diagram, her OCR'ing the image and telling me what and how her own feature-set works
-
-
Much to learn from this 20min session! loved seeing how semantic caching capabilities in #CosmosDB can dramatically improve the performance of LLM based chat App. #Azure #database #LLM #Intelligent #App
🌟 Learn how #AzureCosmosDB powers scalable chat history! Jasmine Greenaway & James Codella show you how with the Microsoft Semantic Kernel. Watch here: https://lnkd.in/gRRJQy2E #NoSQL
Build Scalable Chat History and Conversational Memory into LLM Apps
https://www.youtube.com/
-
🌟 Learn how #AzureCosmosDB powers scalable chat history! Jasmine Greenaway & James Codella show you how with the Microsoft Semantic Kernel. Watch here: https://lnkd.in/eepkMgui #NoSQL
Build Scalable Chat History and Conversational Memory into LLM Apps
https://www.youtube.com/
-
🌟 Learn how #AzureCosmosDB powers scalable chat history! Jasmine Greenaway & James Codella show you how with the Microsoft Semantic Kernel. Watch here: https://lnkd.in/e9BUX_Sx #NoSQL
Build Scalable Chat History and Conversational Memory into LLM Apps
https://www.youtube.com/
-
Internal GenAI use case? we got you! We love our developers and want to make sure they have an amazing technical documentation for all of our products and would love to share how we did! 📖 This led our team to develop an AI chatbot that allows direct communication with MongoDB documentation. Take a look at this tutorial to learn how they did this. 👇 https://lnkd.in/dk5QnENW
-
-
1. The secret sauce in Ellipsis (YC W24)'s smart AI code reviews = deep codebase understanding. 2. A happy side effect of deep codebase understanding = analytics on merged pull requests in Slack. 3. Analytics on merged PR's in Slack = happy customers. And you can be a happy customer too. You'll get deep, LLM-powered code reviews on every commit of every PR and nifty notifications in Slack when PR's are merged. Like this one: Start a free 7 day trail at ellipsis.dev
-
-
We love our developers and want to make sure they have an amazing technical documentation for all of our products. 📖 This led our team to develop an AI chatbot that allows direct communication with MongoDB documentation. Take a look at this tutorial to learn how they did this. 👇 https://lnkd.in/gHQcmn5X
-
-
Curious about the database powering #ChatGPT and Microsoft Teams? You may find the answer below! 💡 In this video, you will learn how to: - Get your database ready for AI with Azure Cosmos DB - Solve for real-time data access requirements globally - Automatic scale with partitions keys - Create for copilot-style apps - Build app using vectorized data - Vector indexing & search with Azure Cosmos DB for MongoDB vCore - Run smaller apps serverless - Set maximum throughput thresholds with controlled cost #cosmosdb #azure #ai #appinnovation #MicrosoftHK
What is the database behind ChatGPT?
https://www.youtube.com/
-
On June 13 ... data is the backbone of AI models... here are best practices on how to get and process the data to your models
Building cutting-edge GenAI applications? You don't want to miss this. Join us on June 13 to: ✅ Learn how to build a real-time-contextualized, and trustworthy knowledge base for your GenAI apps ✅ Discover where #datastreaming and #ApacheFlink fit in retrieval-augmented generation (RAG) architecture ✅ Understand the key steps of data augmentation, inference, workflows, and post-processing ✅ See everything come together in a RAG demo built using Confluent, OpenAI, ChatGPT-4, Flink, MongoDB, and D-ID Secure your spot today! ⬇️
Retrieval-Augmented Generation (RAG) for Real-Time Generative AI Webinar | Confluent | Register Now
-
Start your own GenAI chat-bot journey with a #CodeCatalyst Blueprint
How AWS implemented your #genai companion in #codecatalyst
https://www.youtube.com/
Morgan Kidd, Pengfei Sun, Heather McKelvey, Satish K., Aijia Yan, Slack