Karissa Fuller’s Post

Assistants API VS Llama-Index & MongoDB Hackathon: Kick-off and Introduc… In this hackathon, you will build and iterate on an LLM-based application using AI observability to validate the performance of your app.

🚀 The AI RAG LAB24 is officially underway! Some exciting numbers to highlight: ⏳ 24 hours to complete the mission 💼 5 teams with Namirials from all our global offices 👩💻 This edition boasts the highest number of women developers 🏆 Our 5th Namirial hackathon, and the 3rd one dedicated to AI 🎯 This time, the focus is RAG (Retrieval-Augmented Generation) 🔍 1st edition where each team tackles a specific objective 🤝 1st time we're joined by two key tech partners, Amazon Web Services (AWS) and MongoDB, who will support the teams with their advanced architectures and technologies. Good luck to all the brilliant teams pushing boundaries! 💡✨ #Namirial #AIRAGLAB24 #Innovation #hackathon #AI #RAG

I just launched SearchArena🏟️, a fun little project inspired by UC Berkeley's LMSYS Chatbot Arena. Have you ever wondered how different search providers compare? Well, now you can find out! 🕵️♀️🔍SearchArena lets you compare results from Perplexity, You.com, Brave Search, and Tavily side by side. It's like a friendly competition for search engines! 🥇🥈 Why did I build this? I attended a MongoDB hackathon recently and wanted to create something that would encourage people to explore beyond their go-to search tools. John Willis helped me appreciate that there's so much innovation happening in the search space, and I wanted to make it easy for everyone to experience it firsthand. 🌟 The app is straightforward and built with Streamlit and some API magic. ✨ 🛠️Want to give it a try? Check it out on HuggingFace:  https://lnkd.in/gCtAzVNt Big shoutout to Kye G. Gomez for his contributions and awesome open source fully local perplexity example! (that you can find here: https://lnkd.in/gRXaYzHU) 🙌I'd love to hear your thoughts! Have you tried different search providers? Which one's your favorite? Let's chat in the comments! 💬#SearchArena #AI #SearchEngines #OpenSource #MongoDB #Hackathon

🚀 Exciting News from MongoDB! 🚀 We're thrilled to unveil our latest innovation: the integration of Haystack AI by deepset into MongoDB! 🌟 Dive into our new Hugging Face space: Haystack-MongoDB Integration Chat : https://lnkd.in/dpBKREVX. This collaboration leverages the power of MongoDB Atlas’s vector search capabilities and the intelligence of OpenAI's large language models. Our new space is a game-changer, enabling users to experience firsthand the seamless integration and enhanced search functionalities that come with this partnership. Discover how MongoDB and Haystack AI are reshaping the landscape of data search and AI-driven insights. Join us in exploring the future of data handling and AI technology! To learn more : https://lnkd.in/dHzxQp2B #MongoDB #HaystackAI #Deepset #OpenAI #AIIntegration #VectorSearch #Innovation #TechNews

🧠Memory Hackathon in SF with LangChain, New Computer, MongoDB and Anthropic ✨Sign up here: https://lu.ma/taa6ijxt ✨ Long-term memory for LLM applications is one of the most interesting and important - yet also under explored - areas out there. We are THRILLED to be hosting a hackathon on this with an awesome set of partners. 💻 New Computer is building Dot, an intelligent guide, which uses a novel long-term memory system to enable deep personalization. Samantha Whitmore (CEO) was also an early contributor to LangChain 🍥 Anthropic's new Claude-3 models (with long context window and low latency) are perfect for memory intensive applications 🟢 MongoDB's database - with both vectors and general text data - are optimal for exploring different types of memory We will have a great lineup of speakers and hope to make progress in a cutting-edge area. There are limited spots - sign up quick! https://lu.ma/taa6ijxt

🧠Memory Hackathon with LangChain, https://new.computer/, MongoDB and Anthropic ✨Sign up here: https://lu.ma/taa6ijxt ✨ Long-term memory for LLM applications is one of the most interesting and important - yet also under explored - areas out there. We are THRILLED to be hosting a hackathon on this with an awesome set of partners. 💻 New Computer is building Dot, an intelligent guide, which uses a novel long-term memory system to enable deep personalization. Samantha Whitmore (CEO) was also an early contributor to LangChain 🍥 Anthropic's new Claude-3 models (with long context window and low latency) are perfect for memory intensive applications 🟢 MongoDB's database - with both vectors and general text data - are optimal for exploring different types of memory We will have a great lineup of speakers and hope to make progress in a cutting-edge area. There are limited spots - sign up quick! https://lu.ma/taa6ijxt

LAUNCH DAY! 🚀 LlamaCloud by LlamaIndex is here. 🦙🔥 * Managed Ingestion/Retrieval * LlamaParse (proprietary parser able to handle complex PDFs)

In the realm of Language Model (LLM) applications empowered by Reasoning and Generative capabilities (RAG), the utilization of vector databases is becoming increasingly indispensable. These databases, designed to efficiently store and query high-dimensional vectors, play a pivotal role in enhancing the performance and scalability of LLMs in various tasks, from language understanding to generation. Why are vector databases crucial for LLM with RAG? Efficient Representation: LLMs with RAG operate on large-scale vector representations of data, including text, images, and other modalities. Vector databases provide a streamlined mechanism for storing and retrieving these representations, enabling rapid inference and reasoning. Semantic Similarity Search: The ability to search for semantically similar vectors is fundamental in many LLM applications. Vector databases offer specialized indexing and querying techniques tailored for similarity search, facilitating tasks such as information retrieval, recommendation systems, and content understanding. Scalability and Performance: As LLMs continue to evolve, handling vast amounts of data and computation becomes paramount. Vector databases are designed to scale horizontally, accommodating the growing demands of LLM-based applications without compromising performance. Excitingly, MongoDB a leading modern database platform, is stepping into the realm of vector databases through its support in MongoDB Atlas. With this integration, developers working on LLM with RAG projects can leverage the robustness and flexibility of MongoDB alongside the capabilities of vector databases, unlocking new possibilities for innovation and advancement in natural language understanding and generation. As the landscape of LLM applications continues to expand, embracing vector databases like MongoDB's offering in Atlas is not just a choice but a strategic imperative. The synergy between vector databases and LLMs with RAG holds the key to pushing the boundaries of AI-driven solutions, enabling us to tackle complex challenges and drive meaningful progress in various domains. #LLM #RAG #VectorDatabases #MongoDB #MongoDBAtlas #AI #Innovation #NaturalLanguageProcessing #LinkedIn LlamaIndex

RAG can severely improve the accuracy and quality of the responses generated by LLMs. With Gradient’s newly announced Accelerator Block for RAG, we’re making it possible to set up RAG in seconds - removing friction and complexity. Best of all, it’s powered by best-of-breed technologies like LlamaIndex and MongoDB. By Gradient https://lnkd.in/gS2Y7GVU

AI tools and infrastructure. llamaindex cloud is a great step,