LlamaIndex
Talk to us

LlamaIndex Sep 10, 2024

LlamaIndex Newsletter 2024-09-10

Hello, Llama Enthusiasts! 🦙

Welcome to this week's edition of the LlamaIndex newsletter! We're excited to bring you a range of updates and new tools, including our latest tutorials on Auto-Document Retrieval and deploying microservices with llama-deploy, a robust system for improved workflow management. Check out various other interesting use cases utilizing LlamaCloud and LlamaIndex tools.

If you haven't explored LlamaCloud yet, make sure to sign up and get in touch with us to discuss your specific enterprise use case.

🤩 The highlights:

  • Auto-Document Retrieval Guide: Tutorial on how to retrieve entire documents using RAG and structured querying with a two-pass approach for precise document selection. Notebook, Tweet.
  • Llama-Deploy Launched: Introducing llama-deploy, our latest system for deploying microservices that supports seamless workflow deployments with scalability, fault tolerance, state management, and asynchronous execution. Blogpost, Tweet.
  • Agentic Chatbot Deployment Guide: A step-by-step guide on building responsive chatbots using llama-deploy and Reflex. Code, Tweet.
  • Netchex's Improved HR Support: Netchex's AskHR + Netchex AI, built with LlamaIndex, boosts HR productivity and engagement through advanced RAG pipelines. Blogpost.
  • Multi-Agent Workflow Template in create-llama: A new template in create-llama supports three agents—a researcher, a writer, and a reviewer—with three operational modes: choreography, orchestrator, and explicit workflow, to streamline content creation processes. Code, Tweet.

🗺️ LlamaCloud And LlamaParse:

  • Guide to Auto-Document Retrieval: This tutorial shows how to efficiently retrieve entire documents using RAG and structured querying with a two-pass approach for precise document selection. Notebook, Tweet.

✨ Framework:

  1. We have launched llama-deploy: Our new system for deploying microservices based on LlamaIndex Workflows, offering seamless deployment of workflows, easy scaling, fault tolerance with retry mechanisms, robust state management, and asynchronous code execution. Blogpost, Tweet.
  2. Guide to Deploying an Agentic Chatbot with llama-deploy: This tutorial demonstrates how to use Reflex and LlamaIndex to build a chatbot system that answers users queries. Code, Tweet.
  3. We have added a multi-agent workflow template to Create-Llama which includes includes three agents—a researcher, a writer, and a reviewer—and offers three operational modes: choreography, orchestrator, and explicit workflow, to streamline content creation. Code, Tweet.
  4. Guide to Advanced Agentic RAG Pipelines with Amazon Bedrock and LlamaIndex demonstrating query routing, complex question decomposition, and stateful agent integration for improved reasoning and document parsing. Blogpost, Tweet.

💻 Use-cases:

  • Netchex's AskHR + Netchex AI, built with LlamaIndex enhances employee support with advanced RAG pipelines, boosting HR productivity and engagement. Blogpost.
  • PowerPoint Generation App: Laurie's project with LlamaIndex TypeScript (LITS) transforms speaker notes into a PowerPoint deck, summarizing content, generating slide previews, enhancing aesthetics automatically, and offering a one-click download of the .pptx file. Project, Code.
  • AutoMetaRAG is Darshil Modi's project that optimizes RAG indexing with dynamic metadata schemas using LLMs to adapt to specific dataset characteristics, improving both macro and micro-level data views for better retrieval precision and recall. Code.

✍️ Community:

  • Hanane Dupouy’s tutorial on evaluating RAG pipelines for financial reports with LlamaIndex and Giskard AI, focusing on test generation, component assessment, and clustering.
  • Pavan Kumar’s tutorial on Automating Financial Workflows using LlamaIndex and Qdrant.
  • Karan Vaidya's tutorial to create a personal email and calendar scheduling assistant using LlamaIndex and ComposioHQ's comprehensive email and calendar tools.
  • Rohan’s video tutorial demonstrates how to automate RAG pipelines in LlamaIndex, featuring structured extraction with LlamaExtract, vector indexing, and advanced auto-retrieval querying.
  • Ravi Theja’s video tutorial on building multi-step query engine using workflows.

🎤 Guest Lecture:

  • Jerry Liu will be talking about LlamaCloud and Multimodal RAG at an upcoming Search For RAG online course on September 18th. Check out the course here.

👥 Recruitment: