- unwind ai
- Posts
- Build & Deploy Multi-Agent LLM Apps
Build & Deploy Multi-Agent LLM Apps
PLUS: Opensource and free Bloomberg Terminal, Full-stack AI apps from one prompt
Shubham Saboo & Gargi Gupta
October 08, 2024
Bet you're tired of conferences that are just panels and monologues! That’s exactly why The Product Folks created (Un)Conference!
Join over 350+ visionary leaders, including 50+ founders and CXOs, in Bangalore for India's premier product event. Dive into the future of product development in an AI-driven world, with unfiltered insights and behind-the-scenes stories from founders of InMobi, Clevertap & CXOs from Swiggy, MakeMyTrip, Myntra, Cleartrip, and many more companies.
Don’t miss this exclusive opportunity to learn from the brightest minds shaping the Indian startup ecosystem!
💡 What’s on Deck at (Un)Conference '24?
Cutting-Edge Techniques: Master AI and product strategies driving tomorrow.
Real-World Successes: Hear directly from leaders turning visions into reality.
Make New Friends: Make real connections with like-minded people.
📅 Date: October 19, 2024
📍 Place: Bangalore
🎨 Theme: Building Products for the AI World
🎟️ Act Fast — Spots Are Limited! Readers of Unwind AI get an exclusive discount by signing up with our referral code ‘UnwindxTPF’
Today’s top AI Highlights:
Free and opensource alternative to Bloomberg Terminal
Build, optimize, and deploy multi-agent LLM apps at scale
Google Chrome launches AI writing APIs that work locally in your browser
Opensource Claude Artifacts and v0 in a single AI template to build full-stack AI apps
Run vision language models locally on your Mac using MLX
& so much more!
Read time: 3 mins
AI Tutorials
Building AI tools that can handle customer interactions while retaining context is becoming increasingly important for modern applications.
In this tutorial, we’ll show you how to create a powerful AI customer support agent using GPT-4o, with memory capabilities to recall previous interactions.
The AI assistant’s memory will be managed using Mem0 with Qdrant as the vector store. The assistant will handle customer queries while maintaining a persistent memory of interactions, making the experience seamless and more intelligent.
We share hands-on tutorials like this 2-3 times a week, designed to help you stay ahead in the world of AI. If you're serious about levelling up your AI skills and staying ahead of the curve, subscribe now and be the first to access our latest tutorials.
🎁 Bonus worth $50 💵
Latest Developments
Bloomberg costs $30,000 annually! Tired of being priced out of Wall Street's tools? Here’s a free, open-source financial terminal that serves as an alternative to the Bloomberg Terminal. OpenBB’s Terminal makes advanced financial tools accessible to individual investors, small firms, and developers. The platform combines data access from 400+ sources, data integration, an AI Copilot and more, so you can easily analyze markets, build custom dashboards, and run detailed financial research. Your data is private and you have the flexibility to integrate personal datasets.
Key Highlights:
Cost-Free Alternative - OpenBB offers the same features typically found in Bloomberg, including access to 405+ financial data sources, AI-driven insights, and advanced analytics – all for free.
AI Copilot - The AI Copilot can answer complex queries, summarize financial news, provide trends on stocks or commodities, and assist in generating comprehensive reports. It can also be integrated with private datasets for organizations to use their structured and unstructured data.
Custom Dashboards - You can create fully customizable dashboards, integrate personal datasets (from PDFs, emails, or APIs), and use advanced charting tools to analyze real-time financial trends and data, providing total flexibility for unique workflows.
Exclusive Offer Until Nov 7 - Sign up before November 7 to get free access to premium data sources, including global financial data, economic reports, and market news, for a full year – perfect for analysts, quants, and developers working with financial models.
Build sophisticated multi-agent LLM applications with ease using LazyLLM, a low-code development tool designed to streamline the entire process from initial prototype to optimized deployment. LazyLLM handles the heavy lifting, allowing you to focus on crafting effective algorithms and perfecting your application logic rather than getting bogged down in complex engineering details. Its intuitive workflow facilitates rapid prototyping, data-driven feedback analysis, and iterative refinement, empowering you to create high-performing AI solutions quickly and efficiently.
Key Highlights:
Component-based Architecture - Build applications by assembling functional modules like Lego bricks using pre-built components, pipelines, parallel flows, and other control structures. This modular approach simplifies the integration of various LLMs, embeddings, and other AI services.
One-click Deployment - Deploy multi-agent applications across different environments (bare metal, Slurm clusters, public clouds) with a single click. LazyLLM handles the complexities of starting submodules, configuring URLs, and packaging images for deployment using Kubernetes.
Automated Optimization - Streamline hyperparameter tuning with integrated grid search functionality. Experiment with various base models, retrieval strategies, and parameters without modifying application code. LazyLLM also automates selecting appropriate fine-tuning frameworks and model splitting strategies based on the specific scenario.
Unified Interface for Local and Online Models - Leverage a consistent prompt structure for both locally deployed models (using frameworks like LightLLM and vLLM) and online services (such as OpenAI, SenseNova, and others). Easily switch between local and online models based on project needs.
Quick Bites
Google Chrome is testing new AI browser APIs, powered by Gemini Nano, that work directly in your browser for fast, private results. Early previews include Writer and Rewriter APIs for generating and refining content locally, with more tools like Summarizer and Translation APIs coming soon.
AI video editing platform Captions has introduced Social Studio which employs AI to run your social media accounts. With just a link to your website, the AI will scan your website and plan your content calendar, generate videos featuring you or an AI Creator, and post across all of your social accounts, everything with minimal involvement.
Cohere has released upgraded versions of its Command R and Command R+ AI models, optimized for enterprise RAG. The new versions come with improved efficiency, reduced API price, and enhanced coding, math, reasoning, and structured data analysis.
Tools of the Trade
Fragments by E2B: The open-source Next.js template to build AI-generated apps, integrating E2B Sandbox and Code Interpreter SDK. It supports multiple frameworks (like Python, Vue.js) and LLM providers (OpenAI, Anthropic, Ollama).
Pandas AI: A Python tool that lets you query and analyze your data using natural language, simplifying tasks for both technical and non-technical users. It can be used in Jupyter notebooks, streamlit apps, or as a REST API for more extensive data interactions.
MLX-VLM: Run vision language models locally on your Mac using the MLX framework. It supports command-line inference, Gradio-based chat UI, and scripting for image analysis with pre-trained models.
Awesome LLM Apps: Build awesome LLM apps using RAG to interact with data sources like GitHub, Gmail, PDFs, and YouTube videos through simple text. These apps will let you retrieve information, engage in chat, and extract insights directly from content on these platforms.
Hot Takes
Not fully sure why all the LLMs sound about the same - over-using lists, delving into “multifaceted” issues, over-offering to assist further, about same length responses, etc. Not something I had predicted at first because of many independent companies doing the fine-tuning. ~
Andrej KarpathyThe pro move for Google would be to fully open source notebooklm on GitHub tmrw
Would do 100x more for Gemini developer adoption than literally any of their current platform marketing ~
Anjney Midha
Meme of the Day
That’s all for today! See you tomorrow with more such AI-filled content.
🎁 Bonus worth $50 💵
Share this newsletter on your social channels and tag Unwind AI (X, LinkedIn, Threads, Facebook) to get AI resource pack worth $50 for FREE. Valid for limited time only!
PS: We curate this AI newsletter every day for FREE, your support is what keeps us going. If you find value in what you read, share it wtith at least one, two (or 20) of your friends 😉
Reply