PostgresML

🚀 Implement semantic search in #Postgres in just 15 minutes. Pure SQL, no ML expertise needed. 🐘 Use pgml and pgvector extensions 🔢 Convert text to embeddings ⚖️ Measure vector similarity 📈 Optimize with HNSW indexes 👇 dlvr.it/TFZyVt #AIsearch #databasedev

@rauchg @postgresml has been there before 😎

How to build a better search engine for @ycombinator job listings 🧠💻 Our guide shows you how to: 1⃣ Extract structured data from job listings 2⃣Create semantic embeddings for intelligent matching 3⃣Perform context-aware searches beyond simple keywords Powered by @runtrellis & the @postgresml SDK, Korvus 👇Tutorial: postgresml.org/blog/korvus-tr…

PostgresML now supports Llama 3.2 🦙 🦉 Free for all Serverless Database users 🐘 Bring models to your data, not vice versa ✨ Smaller models, bigger capabilities Read on in the blog: [dlvr.it/TDlprd dlvr.it/TDlprh

Ever wondered how RAG works inside a database? Check out our latest chatbot ➡️ It performs RAG over all of Wikipedia, using open-source models directly in Postgres. ✨Full RAG workflow in one SQL query 🐘In-database processing 🌎100% open-source Try it: postgresml.org/chatbot What do you think about in-database ML? #chatbots #RAG #OpenSourceAI

@parker_lackey 👏 we love to see it. Cheers to more projects that move the models to where the data lives instead of the other way around.

My last post about quackML got some traction so I'm going to go into more detail about it here. I'm making this extension because I believe there is little reason to separate models from the underlying data engine. The AI/ML ecosystem is very scattered. Each language needs its own library or bindings to work with different models. QuackML simplifies this by bringing model management into duckDB directly. Now any language can embed AI/ML simply by calling into a duckDB client in-process. This is still a work in progress but here is a basic roadmap inspired by @postgresml 1. Model training on any data accessible in a duckDB table. 2. Regression, classification, clustering, and GenAI tasks. 3. AutoML by grid and random hyperparameter search. 4. Linear/Logistic regression/classification, lightgbm, xgboost, and huggingface transformer support. 5. Intergration with Lance for RAG based flows. 6. WASM support for training and inference at the edge. Follow along at github.com/parkerdgabel/q… #duckdb #duckDB #MachineLearning

Get all the details on the blog ⬇️ postgresml.org/blog/announcin…

🚀 postgresml-django Seamlessly integrate ML/AI with #django ORM ✨ Automatic in-database embeddings 🔍 Vector similarity search 🧠 ML-powered Django projects Try it now, let us know what you think. github.com/postgresml/pos…

PostgresML is an excellent AI extension to your pSQL database for training, using, and connecting AI 40x faster.

Check out what @sudowrite pulled off with their #RAG stack using PostgresML: ⚡️ Whipped up a prototype in just 3 hours 📈Now crunching 1M+ calls/hour like it's nothing 🐘 Doing all this fancy AI stuff inside their database Get the details here 👉 [postgresml.org/blog/sudowrite…]

@code_star PostgresML / Korvus

Explore the groundbreaking Korvus project, a game-changer for building RAG workflows! This open-source tool leverages PostgresML for in-database machine learning, offering a unified, efficient approach. Learn more: ift.tt/ysPAOro

@Naveen_AIGeek @firecrawl Let us know what you think @Naveen_AIGeek!

@firecrawl @postgresml Integrating Firecrawl with Korvus SDK for RAG in one query is groundbreaking. Can't wait to dive into this tech marvel!

Want to do RAG on websites all with one DB call? The team at @postgresml posted an amazing blog on how to use Firecrawl with their Korvus SDK which unifies the entire RAG pipeline in a single database query. Check it out below 👇

Too many enterprise AI/ML apps are out of date before they launch, and unmaintainable long term. 🙃 Why? Complex data pipelines. X-functional teams work across multiple quarters to get it right. Isn't there a better way? Read on in the blog. ⬇️ dlvr.it/TBd1P9