Miran Bunderla’s Post

I've published a follow-up blog to my previous work on a RAG-enabled system to improved NL2SQL generation for structured data retrieval. In this new post, a discuss a few additional prompting techniques which can further increase the accuracy of generating complex JOINs over the data. https://lnkd.in/dFkU7JMM

Our multi-model database has many built-in functions to handle common tasks, including mathematical functions you can use to detect outliers in your data. Explore the list in our docs. 👉 https://sdb.li/46DJ7DT

Difference Between View and Materialized View If you’ve ever wondered: Why use a View when we have Materialized Views? Which one is better for real-time queries vs performance-heavy reports? I just uploaded a detailed video with practical SQL examples on my YouTube channel. https://lnkd.in/gfnn8_zn

📣 New blog post by Daniël ten Wolde! 🕸️ In this post, we show how to use DuckDB and the DuckPGQ community extension to analyze financial data for fraudulent patterns with the SQL/PGQ graph syntax that's part of SQL:2023. 📖 Visit https://lnkd.in/dBP_KMTE to read the post.

Replacing complex SQL with something graph-optimised always feels like a magic trick. In this case Daniël ten Wolde uses SQL/PGQ as per SQL:2023 to go from "monstrous" recursion to something short and sweet.

The only snapshot you need, to know about data formats while tuning LLMs and whether to use preference tuning or supervised fine-tuning. credits: https://lnkd.in/gqdVBNqS

while i'm searching over web for answer and clarification for this question >>"can i set clustered index on non primary key column in sql database ?" i got this info " *The primary key is a logical construct, designed to uniquely identify each row in your table. That's why it has to be unique and non-null. *The clustering index is a physical construct that will (initially) phyiscally sort your data by the clustering key and arrange the SQL Server pages accordingly. " please check this stackoverflow thread for full answer and clarification. https://lnkd.in/dp8XnTyw

We used to see SQL injection and XSS everywhere. Now prompt injection is becoming common, yet it can be tricky to show its impact. Adversaries are getting creative in how they exploit this class of vulnerability 🤓 😈 ; from data theft and leaking system prompts, to poisoning datasets and manipulating outputs. I will break down some real-world scenarios in my next blog. Stay tuned 😁

Ever run into a divide-by-zero error? It usually breaks your query until you add a clunky 𝐂𝐀𝐒𝐄 𝐖𝐇𝐄𝐍. There’s a cleaner way: 𝐍𝐔𝐋𝐋𝐈𝐅(). 👉 What it does: compares two expressions and returns 𝐍𝐔𝐋𝐋 if they are equal; otherwise, returns the first expression. Example: You’re calculating conversion rates, but sometimes total_visits = 0. Instead of breaking the query, 𝐍𝐔𝐋𝐋𝐈𝐅(total_visits, 0) safely returns 𝐍𝐔𝐋𝐋, and your query runs smoothly. 👉 It’s a tiny function, but it keeps your calculations clean and error-free. #dataanlysis #sql

Day 13 of my SQL Q&A Series Question: What’s the difference between CHAR and VARCHAR? Answer: CHAR: Fixed length → always uses full defined space. VARCHAR: Variable length → saves space dynamically. Example: CHAR(5) always stores 5 characters, while VARCHAR(5) stores only what’s needed. Tip: Use CHAR for fixed data (like codes), and VARCHAR for variable data (like names). #SQL #DatabaseDesign #SQLTips #DataAnalytics #DataScience #SQLLearning #FenilPatel #DailyLearning