Tampa Bay Generative AI

Build an automated resume insights agent - powered by core parsing, extraction, and structured output modules ???? This blog by Fermin Blanco is a fantastic tutorial on building on building a practical example of AI in recruiting: given any unstructured resume, automatically extract out relevant information from it, and then return insights in a structured output format (that you can easily plug into a workplace application). Powered by LlamaIndex, LlamaParse, and structured output capabilities with Gemini. https://lnkd.in/gqyxZMWk

Build a chat UI for your LLM app in minutes with LlamaIndex chat-ui! ???? This React component library offers: ?? Pre-built chat components (message bubbles, input fields) ?? Tailwind CSS customization ?? Easy integration with LLM backends like Vercel AI Key features: ? Custom widgets to extend components ? Code and LaTeX styling with highlight.js and KaTeX ? PDF viewer integration Learn more and start building: https://lnkd.in/ghNG6hZT

ICYMI: You can literally spin up a full-stack financial analyst to fill in a CSV/Excel using any context (e.g. 10Ks), in one line of code ????. Fully open-source ?? create-llama is the superweapon that spins up a full-stack multi-agent application with full API services, data models, and nextjs frontend - you can use it out of the box or customize the code for whatever you wish. Powered by LlamaIndex workflows, and complete with streaming and UX elements around spreadsheets. Huge shoutout to Marcus Schiesser. Check out create-llama: https://lnkd.in/gPTYt6um

?? ????????????-?????????? ???? Hugging Face solved one of AI's biggest headaches - the "bigger means better" problem. Their new SmolLM2 is a series of surprisingly powerful small language models that can run right on your device. Key highlights: > Uses just 1.7B parameters (compared to GPT-4's hundreds of billions) > Actually outperforms Meta's larger models in some tests > Runs directly on your phone or laptop > Zero cloud dependency needed This could democratize AI development. Is this the future of practical AI?? Share your thoughts - especially if you're working on implementing AI in your business. Check out the Model Series here:https://lnkd.in/df7xSMMB #AI #MachineLearning #ArtificialIntelligence?

Reliable Text-to-SQL over 500 Tables ???? Most text-to-SQL tutorials are easy and operate over “trivial” examples like 2-3 tables. This tutorial by Ryan Nguyen is one of the best we’ve seen in showing you how to construct a SQL agent that can operate over a large and complex data model (500+ tables, relationships between them). Here are some of the key steps: 1. Iterate through each table and extract a structured schema with LLM-generated summaries 2. Use hierarchical chunking and indexing to help retrieve an initial set of relevant tables 3. Use graphRAG techniques to provide tables related to the existing tables Feed relevant and related set of tables to text-to-SQL prompt to generate the query https://lnkd.in/guvp7nkQ

Since the launch of ChatGPT almost two years ago Nvidia, Microsoft, Meta, Apple, Amazon and Alphabet have added $8.2 trillion in market cap ?? Nvidia alone has added $3T in market cap since the launch of ChatGPT. Today, Nvidia accounts for over 9% of the Nasdaq index, just behind Apple at 10%. In total, tech's big six account for more than 40% of the Nasdaq index. Never has the index been this concentrated. Almost all of the gains in the tech heavy Nasdaq during the past 24 months come from big six tech stocks below. Very few saw this coming in 2022 when we we're heading into a recession. #nasdaq #artificialintelligence

?? There’s been a lot of talk about LAMs (Large Action Models) recently. What are they, and how do they differ from LLMs? Here’s some scoop :) ? Large Action Models (LAMs) are AI models that go beyond understanding and generating information to perform complex actions. ? LAMs are designed to handle multi-step problems and operate on a large scale enabling advanced autonomous decision-making and task execution. ?? Wait, isn’t this the same as AI agents? How are they different? ? AI agents focus on making decisions based on existing data and user interactions. LAMs, on the other hand, are designed to handle more complex, multi-step tasks by learning from massive, action-oriented datasets. ? LAMs don’t just interact with data; they optimize sequences of actions, pushing the boundaries of autonomous decision-making and execution on a much larger scale than traditional AI agents. ? Think of LAMs as integrated agents that use deep learning to evolve and learn from their actions, rather than being simply engineered through LLMs and predefined connections. I see them more as integrated, evolving AI agents that interact with their environment and continuously adapt as they complete tasks. ?? Honestly, I think we're overcomplicating things with all these new names, but I do see the subtle differences and why a new term might be justified, at least in this case. Image Source: https://lnkd.in/er34_ubB (Love the article too, do read it if you have the time!)

The process of building Multi-Agent #RAG Systems with #LlamaIndex involves several key steps. It begins with data ingestion, where various information sources are input into the system. This is followed by index creation, where LlamaIndex efficiently organizes the data for quick retrieval. The next step is query engine setup, configuring the system to effectively search and retrieve relevant information from the index. Simultaneously, agent definition and task planning occur. Agent definition involves specifying the roles and capabilities of different AI agents in the system, while task planning breaks down complex queries or tasks into manageable subtasks for these agents. Once set up, the system moves to agent execution, where multiple AI agents work on their assigned tasks, leveraging the RAG system to access and process information as needed. Finally, result aggregation takes place, combining and synthesizing the outputs from various agents to produce a cohesive final result. This workflow enables the creation of sophisticated multi-agent systems that can handle complex, information-intensive tasks by combining the strengths of retrieval-augmented generation with collaborative AI agents. Here is a complete talk by Jerry Liu on Building Multi-Agent RAG Systems with LlamaIndex: https://lnkd.in/gmGcX2T5

Which framework works best for you?

??Open Canvas Open Canvas is an open source web application for collaborating with agents to better write documents. It is inspired by OpenAI's "Canvas", but with a few key differences: ??Open Source ??Built in memory ??Start from existing documents ??Open Source: All the code, from the frontend, to the content generation agent, to the reflection agent is open source and MIT licensed. ??Built in memory: Open Canvas ships out of the box with a reflection agent which stores style rules and user insights in a shared memory store. This allows Open Canvas to remember facts about you across sessions. ??Start from existing documents: Open Canvas allows users to start with a blank text, or code editor in the language of their choice, allowing you to start the session with your existing content, instead of being forced to start with a chat interaction. We believe this is an ideal UX because many times you will already have some content to start with, and want to iterate on-top of it. Video: https://lnkd.in/giSWr_ag Code: https://lnkd.in/g4P2esqT Try it out here: open-canvas-lc.vercel.app