Large language models are ever more commonly handling sensitive data at scale. 📈 This has turned the issue of exposing specific data to LLMs into an essential topic of consideration as AI capabilities continue to expand and integrate deeper into applications. RAG Servers and

What's ahead for cloud in 2026? Our Chairman and Chief Product Officer Phillip Merrick shares his predictions in DBTA's latest feature, "9 Predictions for Cloud in 2026," alongside insights from industry leaders at AWS, Oracle, Cloudera, and more. Check out Phillip's take on

https://t.co/QjNnXDCBKp has written a great article on pgEdge’s new open-source MCP Server Toolkit for PostgreSQL. This piece walks through how MCP can help connect Postgres data to AI agents, RAG workflows and other emerging AI tools. Read the full article here: 🔗

Using pgEdge means you get the flexibility of scaling #PostgreSQL from any point, even from a single-primary node, up to multiple nodes spread across multiple regions. Our open-source components give you write-anywhere, read-anywhere active-active capabilities with pure

Trying to set up semantic search for your #PostgreSQL database? If you've ever wanted to understand how to: 👉 Chunk documents into semantically meaningful pieces 👉 Embed vectors for each chunk enabling similarity search 👉 Automatically synchronize when documents change 👉

Excited about the latest news regarding our open-source agentic AI offerings for Postgres? Take part in the discussion on HackerNews - we'd love to hear what you think! Leave feedback on the Agentic AI Toolkit: https://t.co/rQ6WrToEwN Or talk to us about the Postgres MCP server

MCP servers: making it possible for AI agents, including code generators, to access data as well as understand the database schema, relationships between data objects, and performance metrics. Now, you can get an open-source MCP server for #Postgres that can be used

pgEdge Agentic AI Toolkit for Postgres is now available for free to all Postgres users!. Deploy on-premises (even air-gapped), on self-managed cloud, or a managed cloud service via pgEdge Cloud. What's in it? 👉 A highly performant and full featured MCP server 👉

It isn't always easy to get up and running with PostgreSQL or to get the results you expect - especially at scale. At pgEdge, we have a staff of highly trained, dedicated experts with decades of experience working with Postgres - and we're here to support you (from

If you've ever wanted to build a Retrieval Augmented Generation (RAG) server with #PostgreSQL, now's your chance. Dave Page wrote a 3 part series using OpenAI, Voyage AI, or a local Ollama installation. However, he does note: > Whilst it's not supported, I did play with

Protecting personally identifiable information (PII) within databases has become an integral part of database administration, due to data privacy regulations such as GDPR, CCPA, and HIPAA. If you've ever wondered if there could be an easier way to anonymize sensitive data as a

The first #PostgreSQL #Edinburgh meetup was a success! Did you miss @Jimmy Angelakos delivering RAGTime with Postgres? His comprehensive slides on learning about AI power with pgvector and Retrieval-Augmented Generation can be found here: https://t.co/qjjbkbVIcq cc @Postgres

In our latest video, Solution Engineer Paul Rothrock demonstrates how pgEdge's Zero Downtime Add Node feature within Spock enables seamless cluster scaling and major version upgrades—without any service interruption. (Scripts to automate this process are available here:

Zero-downtime PostgreSQL maintenance is now real with pgEdge. Upgrade, scale, and add nodes without taking your DB offline. 👉 Read the blog: https://t.co/yzG8Zuia43 #PostgreSQL #HighAvailability #ZeroDowntime #OpenSource #pgEdge

In the final post of the "Building a RAG Server with PostgreSQL" series, learn how to deploy the open-source pgEdge RAG Server for a simple HTTP API you can use to ask questions about your content, and get answers based on your own documentation. Part 3 is now live, check it out

Why did we create snowflake sequences when UUIDs exist? ❄️ While they're very similar in purpose, snowflake sequences are a little more efficient because they encode information like the creation timestamp and the ID of the server that created it while being half the size of a

Did you know the b-tree skip scan capability was added as a query optimization improvement in Postgres 18? ✨ This allows the query planner to use multi-column indexes even when early columns lack equality restrictions - something that's especially useful for analytics and

Local to #Edinburgh? The next #PostgreSQL Edinburgh Meetup is this Thursday, December 11 and you'll be able to catch @Jimmy Angelakos delivering "RAGTime with Postgres: AI Power with pgvector and Retrieval-Augmented Generation". It's bound to be a great deal of fun - register

pgEdge Vectorizer: an open-source #PostgreSQL extension that automatically chunks text content and generates vector embeddings using background workers. It makes it easier to build AI-powered search and retrieval applications asynchronously. If you've never used the extension

Now, you can use open-source pgEdge components designed for high-availability & multi-master replication combined with #CloudNativePG for cloud-native #PostgreSQL orchestration. Our goal was to make it easy for developers to adopt powerful open-source and vendor lock-in free