Can't wait for this series to go live! Had a blast recording this!! I will walk through how to build AI Agents with your Postgres Database like a Pro! #postgres #azure #ai #agents

Excited to be part of this PostgreSQL on Azure series! I’ll be sharing a video on Scaling to meet the most demanding workloads focused on practical ways to handle scale, performance, and real-world challenges. Check out the full series below 👇 #PostgreSQL #AzureDatabaseForPostgreSQL #Scalability #Cloud

We’ve been working on a new series with the Azure Database for PostgreSQL, Azure HorizonDB and AMD teams 🎉. The new PostgreSQL Like a Pro video series designed to help you unlock the full potential of PostgreSQL across the Microsoft ecosystem, from Azure and VS Code to AI-powered and agent-driven apps. PostgreSQL is already an amazing foundation, but stitching together performance, scaling, migrations, and AI experiences can be tricky. In this series, we focus on making that journey smoother with practical guidance and real demos you can actually use. Here’s a quick preview of what we cover: • Building smarter AI agents with MCP and vector search • Simplifying migrations with AI-assisted tooling in VS Code • Optimizing performance and resiliency on Azure • Scaling with new capabilities like Azure HorizonDB our new PostgreSQL based offering. If you’re building with PostgreSQL and want to learn more, this series is for you. Watch the series: https://lnkd.in/e5DEHMen Read our blog: https://lnkd.in/eqfsBwnq

[Azure Updates] [Launched] Generally Available: GitHub Copilot in SQL Server Management Studio 22. GitHub Copilot is now generally available in SQL Server Management Studio (SSMS) 22, bringing AI‑powered assistance directly into the application that you use every day. With GitHub Copilot in SSMS, you can use natural language to develop, explain, fix, #azure #azureupdates https://lnkd.in/ezmxpmrA

🤖 Would you like to enable your AI agents to safely interact with SQL databases using Model Context Protocol. 🎉 Then you might be happy to hear that the official Microsoft SQL MCP Server now is available for public preview. ❓This new SQL MCP Server is a feature rich component of data API builder (DAB) that gives you a simple, predictable, secure way to bring AI agents into your data workflows without compromising the database or relying on natural language. 🏠 The MCP server can run both locally, on-premises, or in any cloud as a containerized service. 👇 Click the link below to learn more: https://lnkd.in/e_thKFzM

🚀 Azure Databricks makes data feel effortless—VS Code feels like heavy lifting. Perks of spending years in DBX. No more going back to VSCode.

Server down!! Server down!! Good evening humans of earth. As you may know, I have been on a mission to get all our data platform stuff to SaaS because infra is non when you're running a small team with extensive scope and limited resources. It was very satisfying to kill the legacy SQL Server in 2023 (ala Arnie: "consider that a divorce"). And that dumb SFTP server, well good riddance to that too. But there was some persistent EC2 footprint and the final instance is ready to be powered down. The last push was to get a process that isn't meant to exist, and nobody really wants or knows about, off that platform. I got AI to write some python scripts and rehouse to AWS lambda. If AI never does anything else, helping me avoid suffering with boto3 is enough. So, come resumption of business next week, the server will be down, and it won't be a sev1, it will be...Zen.

Interesting that Acalanes High School has created some of the world's best leaders in "AI" development. Bret Taylor has been innovating for years on some of the world's most successful platforms. Today, he leads Sierra, a fast-growing team supporting CX and CS. Leverage multiple channels (chat, email, text). Then there is Shehab Amin and Everett Roeth, who have developed the number one alternative to Apache Spark that focuses on doing more with less. Lakesail.com 1. Sail (by LakeSail) — the strongest current drop-in replacement • Description: A Rust-native distributed compute engine that implements the Spark Connect protocol. • This means you can run existing PySpark, Spark SQL, or DataFrame code with little/no changes while getting Rust’s speed and efficiency. • Claims: Often 4x faster than traditional Spark, with dramatically lower infrastructure costs (e.g., ~94% less spend in some benchmarks like TPC-H). • Maturity: Actively developed and seeing real-world excitement (HN/LinkedIn discussions in 2025). • Best for: Teams that want to keep Spark APIs/ecosystem but escape JVM limitations. • Repo/Info: Search for “LakeSail Sail” or check http://lakesail.com (open-source Rust engine). What was in the Lafayette water that we didn't get in Orinda? OH, wait, we killed them in sports...

Finally got Apache Airflow running on AWS EC2 — but the journey there was a bit of a mess 😅 Honestly, I was stuck for longer than I'd like to admit. I started with `airflow standalone` — seemed like the simplest way to get things running watching tutorials. But the UI just wouldn't open. And the terminal? Completely stuck. No errors, no clues. Just... nothing. So I thought maybe `airflow webserver` would do it. Tried that. Same story. At that point I was genuinely confused. Everything *looked* like it should work. But nothing was showing up. Finally did what I probably should've done from the start — **read the documentation.** Turns out Airflow 3 works differently. Two specific commands are what actually get the UI up and running: → `airflow api-server` — this is what serves the UI now → `airflow scheduler` — this handles the DAG execution That was it. That was the whole fix. Two commands I didn't know about because I was relying on what I'd seen in old tutorials instead of just reading the docs. The moment the UI loaded in my browser I felt equal parts relieved and slightly embarrassed 😂 **Biggest lesson? When you're stuck — just read the docs. Seriously.😅** Next step → building a real data pipeline with Airflow👍 #ApacheAirflow #AWS #DataEngineering #LearningInPublic #CloudComputing

🚀 AWS Learning Series – Day 8 📘 Topic: S3 Performance Optimization Today I learned that storing data is not enough — how fast you access it matters just as much ⚡ Here’s what stood out 👇 ✔ Use prefixes to handle high request loads ✔ Multipart upload improves large file transfers ✔ CloudFront reduces latency by caching data closer to users 💡 Key Learning: Performance optimization = better user experience Real-world thinking: A slow app = users leave A fast app = users stay This is where AWS really shines 🚀 #AWS #CloudComputing #AWSS3 #LearningInPublic #TechJourney #Performance #CloudFront