PART I: 𝑨 𝒍𝒐𝒏𝒈 𝒕𝒊𝒎𝒆 𝒂𝒈𝒐, 𝒊𝒏 𝒂 𝒈𝒂𝒍𝒂𝒙𝒚 𝒇𝒂𝒓, 𝒇𝒂𝒓 𝒂𝒘𝒂𝒚... 𝒊𝒏 𝒄𝒐𝒔𝒊𝒏𝒆 𝒔𝒊𝒎𝒊𝒍𝒂𝒓𝒊𝒕𝒚. ���️
This Star Wars Day, join us on an epic journey where vector search brings balance to the force of unstructured data! Just as the Jedi use the Force to connect all living things, Zilliz's vector database technology connects and makes sense of all your data across the galaxy of information. 🌌
May the 4th be with you as you explore new dimensions in vector space!
#MayThe4thBeWithYou #StarWarsDay #VectorDatabases #Zilliz #Milvus
Write SQL with @vannadotai and #Milvus 🤝
Last week, community member Deepak Jhanji presented a solution using Vanna AI and Milvus: A bot that generates SQL queries from natural language inputs, executes them, and provides results while allowing user feedback for continuous improvement.
🤝 This integration combines Vanna's RAG framework for SQL generation with Milvus, the advanced open-source vector database, creating a powerful tool that bridges natural language and database queries.
How It Works:
1. Train on your database schema, documentation, and example SQL queries
2. Ask questions in plain English
3. Execute the generated SQL against your database
4. Improve through user feedback
Key Benefits:
✅ Makes databases accessible to non-SQL experts
✅ Reduces time writing repetitive queries
✅ Produces reliable, contextually aware SQL
✅ Learns and improves over time
The project structure is simple, using Python connectors for the databases and a clean Streamlit interface. The system correctly interprets business context to generate appropriate queries.
⭐ Project repo: https://t.co/V6cucy3fxx
⚒️ Milvus docs: https://t.co/2Lzq2nIEZF
We made 𝐃𝐞𝐞𝐩𝐒𝐞𝐚𝐫𝐜𝐡𝐞𝐫 to illustrate how easy it is to build a research agent with #Milvus and a reasoning LLM. 🧑🎓
And, it's open source! 🤝
🛠️ Try it out: https://t.co/KwM2zHAQ6G
📚 Learn more: https://t.co/FpidCwkLjO
🔥 Follow us for more tutorials like this
#LLM #DeepResearch #OpenSource
🐉 NEW PODCAST: 𝑺𝒕𝒓𝒖𝒄𝒕���𝒓𝒆𝒅 𝑶𝒖𝒕𝒑𝒖𝒕 𝒘𝒊𝒕𝒉 𝑶𝒖𝒕𝒍𝒊𝒏𝒆𝒔 with @just_cameron from @dottxtai and @stefan_webb 🐉
Cameron shares his journey into the field, the mission of .txt, and the importance of constrained sampling in AI development.
🎥 Watch: https://t.co/NZtLnp2UE0
🎙️ Listen: https://t.co/cPWUbZOKFg
#Podcast #AIPodcast #Developers
There are so many tools for ETL, but which one fits YOUR use case? 🤔
We compared the most popular document processing and ETL tools for RAG, and here are our findings. While they serve different purposes, all integrate with Milvus and Zilliz Cloud. 🤝
Tutorial links to each👇
• @AirbyteHQ: https://t.co/VwA1GOREPd
• @fivetran: https://t.co/HXqRnVEyWx
• @UnstructuredIO: https://t.co/ax09jECu0d
• @Vectorizeio: https://t.co/0D1jha04Hz
• @GetUnstract: https://t.co/HhjCJmWH3e
#UnstructuredData #AIEngineering #ETLComparison #VectorDatabases
Happening this week! 🌟 Unstructured Data ~Workshop~ with #Milvus and @OpenAI is this Thursday, March 20 in Palo Alto. Join @stefan_webb and Harry Zhao (OpenAI Engineer) to learn how to amplify AI with agentic workflows. 🔥
Save your spot now! 👉 https://t.co/ci6rosd3MM
#Workshop #OpenAI #Vectordb #Agents
1/ 🧵 We open-sourced DeepSearcher: A Python-based research agent that lets you build LLM-powered research tools locally. Here's the technical deep-dive…
#Python#OpenSource#DeepResearch#DeveloperTools
🔥 Build RAG pipelines with @firecrawl + Milvus: Transform Your AI Applications!
Easily convert website content into clean, structured data ready for AI workflows.
✨ How It Works:
1️⃣ Scrape and Structure: Use Firecrawl to extract clean, structured data from websites.
2️⃣ Embed and Store: Convert data into embeddings and index them in Milvus for high-speed retrieval.
3️⃣ Ask and Retrieve: Build your RAG system to retrieve relevant data and supercharge your LLM’s responses.
Full link 👉: https://t.co/L4krPdRmsK
#Milvus #Firecrawl #GenAI #RAG #WebScraping #AI
🔥 Milvus & @OpenAI are teaming up for an in-person AI workshop! 🔥
📅 March 20, 2025 | 5:30 PM
📍 Palo Alto
💡 Talks + HANDS-ON AI agent-building!
🤯 Learn how O1 → O3 reasoning changed the AI game
🛠 Build an AI agent that reasons over real data
🎟️Register now! https://t.co/ci6rosd3MM
🍕 Food, drinks & top AI networking included!
#AI #OpenAI #Milvus #MachineLearning #AIagents #TechEvents
A new podcast episode is out: Prompt, Score, Repeat: Principled RAG & Agent Design🤖
@stephenbtl and @stefan_webb explore the increasingly important world of GenAI model evaluation.
Links to watch or listen are below!
#AIPodcast#RAG#Evaluation#GenAI
FULL Lineup 🔥for the next SF Unstructured Data meetup on Feb 26 includes talks from @awscloud, @HasuraHQ, and @zilliz_universe. Learn about hybrid search, data automation, and achieving 100% accuracy in AI systems.
Save your spot: https://t.co/c2sFa7u1NS
#Meetup#SanFrancisco #TechEvent
Learn all about Lexical Search from @stefan_webb at our Unstructured Data Meetup on Wednesday, February 26! ✨ He'll be covering:
➼ Why, when, and how should you use lexical search
➼ What is the BM25 distance metric
➼ How exactly does Milvus 2.5 implement lexical search
➼ How to build an improved hybrid lexical + semantic search with Milvus 2.5
Save Your Spot: https://t.co/c2sFa7ttYk
#Meetup #SanFrancisco #AIEvent #UnstructuredData #Vectordatabase
🛍️ Weekend project: Multimodal Product Recommender Demo Using #Milvus and @streamlit
Ever wished you could find products by showing a picture and describing what you want? Run this Milvus #multimodal#RAG demo to create a product recommender!
Try it: https://t.co/rlplC5fDMh
#Multimodal #RAG #Demo #Milvus #VectorDB
ICYMI -@yi_ding spoke about advanced #RAG techniques at our South Bay Unstructured Data Meetup! 💪 Small to Slide RAG enhances the performance of multimodal LLMs when dealing with visual data like presentations or documents with images
Read the recap: https://t.co/UAOO1p09I8
Introducing Milvus 2.5 with built-in full-text search capabilities and more! 🔍 Here’s the TLDR on new updates:
🔎Built-in full-text search with Sparse-BM25
💬Advanced text matching using Tantivy's analyzers & Indexing
🏃New Bitmap indexing for faster filtering
+More!
Links below to learn more 👇
#Milvus #Launch #Vectordatabase #Search #HybridSearch #FullText #ML #AI