Voice of the DBA

Tag: AI

  • Learning a New Language

    This week, there are two interesting articles in the newsletter. The first is a markdown getting started article, which helps you learn how to format your documents. At first, I wasn’t sure I liked markdown, but I’ve come to prefer it for a lot of documents. It’s easy to use, and it makes using Git for docs very easy. Plus, I can mix in markdown and HTML as needed, which I really appreciate. I think most people who collaborate and update any documents in a team should learn to use markdown instead of Word, Notepad, etc.

    The second one that caught me eye is one on Power Query. I don’t know Power Query, but there was a piece on the 13 things John Kerski wishes he knew when he started. I didn’t really do more than skim this since I am not writing any Power Query right now. What was interesting to me is that I’m not sure I’d bookmark this if I were actually writing Power Query in Power BI as part of my job.

    Maybe I would, but as I start to use a GenAI more and more to get me started with different projects or requirements, I’m finding it as a way to kickstart me in learning something new. It’s not perfect, and I have to test the results I get. Learning to judge the quality is part of my job, but that’s different than trying to learn something new from scratch. That’s a similar view to what I’ve heard from a few people. This podcast with Kent Beck showcases this view (starts at 6:12).

    In the real world, when I wanted to learn something new, I often looked for a book or a few magazine articles that might get me some basics in a new area. In the digital world, I used to do the same thing, but in the last 20 years, I’ve been looking for an article or a “start in 30 days” book that might give me some basic skills. Then I could experiment and extend what I learned to get work done, likely asking friends or someone in a forum how to do the thing I need.

    Now I ask a GenAI and go from there. I no longer worry about learning the specifics of a language, as I have a good idea in general of the things I need to do. What I’m using the GenAI to do is implement some details, and then I go back and check things. It’s like having an intern who can do a first draft while I work on something else.

    Except it’s way faster.

    GenAI models aren’t perfect, and they make mistakes. However, the more I use them, the more I enjoy them, and I find enough small time savings that I want to use them more.

    Steve Jones

  • The Next Great Thing

    At SQL Bits, I was chatting with Brent about a few things, including AI, which we think is changing the world. I’ve got my set of AI experiments going, and I do believe we will fundamentally alter work and how we use computers in the future.

    Not sure if it’s for the better or worse, but things are changing and will change more.

    Brent mentioned he was thinking about how a few years ago the “next great thing” was blockchain, which was going to change databases and storage, but never really caught on. I never thought it would as it seemed too niche-y for me.

    What other great waves of tech have you seen over the years? I saw a prediction that data engineering will be dead in 5 years. I’ll take that bet. I’ve been hearing the DBA (or other data related positions) will be gone soon for over 25 years. I remember friends asking me if I was worried about my position after SQL Server v7 was released in 1998. I wasn’t then, and I’m not now. It seems there is more work than ever, though certainly the bar is higher for people doing the work.

    You need to learn to be better at your job if you want some job security. That’s one reason I constantly push people to learn, improve, and show off ( consider #SQLNewBlogger posts) their skills. Employers want to know you’re providing value if they’re paying you $$ (or € or £ or ¥ or whatever).

    There have been plenty of tech flops. Virtual/Extended/Augmented reality and various headsets haven’t quite taken off, though there is no shortage of new “AI powered” eyeglass products. Quantum computing still seems to be only slightly less fantastic than fusion energy. Voice control or gestures were supposed to make keyboards obsolete, but we still type a lot.

    There are also some amazing things. Touch computing, especially on mobile, dominates. More and more digital payments are changing our commerce functions. Smartphones are essential devices for most people with apps available for anything and everything.

    I do think GenAI will continue to change the world and how many of us work. This was an interesting talk on how AI changes work, not from a technologist per se, but from a writer. The GenAI models continue to improve, and I find them most useful in saving me minutes, not doing all my work. They are becoming more helpful, though slightly untrustworthy assistants. They are eager teenagers that can help, but need supervision and guidance.

    And sometimes need to be abandoned for the current task and you need to do the rest of (or all of) the work.

    What memories do you have of past “next great things? or maybe of what you see for the future?

    Steve Jones

    Listen to the podcast at Libsyn, Spotify, or iTunes.

    Note, podcasts are only available for a limited time online.

  • T-SQL Tuesday #189–AI and your Career

    I’m late to the party this month. Taiob Ali has a great invite for a topic that is likely on most people’s minds: AI and your career. I constantly hear people asking about this (well not lately, I’ve been on sabbatical).

    I love the T-SQL Tuesday blog party and hope more people participate. Spread the word, ask others to write, and help promote this on socials.

    AI and My Career

    I’ve been a bit skeptical that AI would really help me. For the last 15-20 months I’ve been using AI in different ways, experimenting with things and seeing where it might be useful. A lot of my use has been in VS Code and Copilot, where I do some coding, and a lot of markdown/HTML management. Across that time I’ve found AI to be more and more useful with reformatting or suggestions completions.

    More importantly, I’ve learned to “see” the hints and suggestions quicker and have AI save me time in little ways. None of the items are big, and it hasn’t built me a big piece of code, but I have found it to be very useful in saving minutes. Saving minutes multiple times a day starts to add up in the week and months.

    My Outlook

    I think AI will help me in my career in two big ways. First, there is the time saving and assistance it gives me that lets me be more productive. I look forward to more agents that can be configured to handle some tasks for me and just get things done. I haven’t been a big Siri/Hey Google person with setting appointments and small tasks, but I can see some of this starting to be more useful over time. I bet I can get some services and agents to do some work for me with natural language that eases my job.

    I also think this GenAI tech (and other AI) will help me learn new skills and techniques. I’ve seen some people talking about using an AI to help them learn and I need to do some experimenting here. I tried to get it to build some web apps for me and it didn’t go as smoothly as I’d like, but I’m anxious to see if I can get it to help me learn how to better code in new ways as well as generate code for me.

    It’s important to be able to judge if the AI is making good decisions, and that requires knowledge. In some cases, I have that knowledge. In others I don’t, but maybe AI can help me learn.

  • Sample Data with GenAI in Seconds

    A friend was asking for help with some data analysis. This was in PowerBI, and the source data was PII. I asked them to mask the data for me to work with but realized that’s a chore. I could mock it myself with AI.

    This is part of a series of experiments with AI systems.

    The Process

    I started by building a table. It’s what I often do to solve problems. I could have asked the AI, but I wasn’t sure what I wanted and across 5 minutes or so I was thinking about the problem and what data was needed.

    This is a subset. The original table and model had many more columns, but after thinking for a couple of minutes, I realized these would let me model the report someone wanted. At least enough to see if the requirements might be correct and the code work or not work.

    CREATE TABLE MarketData
( UserID INT
, ContactDate DATETIME
, ContactType VARCHAR(10)
, ContactDesc VARCHAR(30)
)
GO

    Once I had the code, I jumped over to Claude to give this a try. You can see the prompt below, but it was this: I need sample data for this schema. Duplicate some customerIDs across various dates, and duplicate contacttype for each customer (from 1 to 5), but randomly mix these in across dates. Not every customer needs every contact type. Make this for 10 customers across two years

    2025-06_line0121

    I could then move on to another task, since I was getting going in the am and answering emails around this task. As it generated, I realized this wasn’t quite what I needed. So I addd this prompt.

    2025-06_line0122

    I got a bunch of data.

    2025-06_line0128

    This wasn’t perfect, and I actually went back and asked for more users with only a “1” and got the data altered.

    I could have done this, but I’ve also found myself sometimes spending as much time mocking an environment as I did writing a query to solve a problem. I can use this to answer questions on SQL Server Central and even quickly strip code out of poorly asked questions.

    It’s amazing. Don’t be afraid of AI. Give it a try. Keep trying. If it fails after a few prompts, just do the work you’re do, but go back and try again later.

    I’ve got a video walkthrough of this below.