Technical History of Paige

For a personal blog, I don’t spend much time on here talking about myself. It’s my place to talk shop, even when there’s no one around to talk shop with. But I gotta wonder, why are you readers taking my word for any of it? You have no idea what I know or how I know. So, I figured there ought to be a post about that.

Plus, I’ve been interviewed a few times in the last few weeks, and it’s got me thinking about how my history gives me a unique perspective on the data management industry.

I started out back in the nineties doing tech support and documentation for an ETL (Extract Transfer Load) software startup. I quickly added QA (Quality Assurance, aka testing), documentation and support for a GUI-based semi-structured data parser. I learned how data management worked, what people were trying to do, what went wrong, how to use the software to fix it, and helped design software improvements.

That lead to studying how the software was built, what it was doing behind the GUI (Graphical User Interface) and learning programming languages. Next thing I knew, I was working as a software engineer, fixing bugs that, as a support technician, I’d reported!

Good thing I did a decent job of including bug reproduction instructions.

I did software engineering for a few years, then did consulting. I built data pipelines for people from scratch, using the software I’d helped design, document, support, and build.

Then, I went into marketing, and wasn’t that a culture shock.

Suddenly, I had to stop talking about the technology that fascinated me and speak a whole different language, so executives and line of business people understood what the tech could do for them. I learned how to communicate in ways that really benefitted me when I went back to consulting. Wish I’d known how to bridge that gap before.

Consulting was cool. When you write software, it’s like you’re solving tiny, intricate puzzles every day. When you consult, you’re solving a far bigger puzzle, then solving all the little day-to-day puzzles that get the big one solved in the end. And, you have to communicate what you’re doing and why to folks who have a huge stake in your success, but don’t really understand what you do. That was what I learned from marketing, how to help people understand why one way of solving the puzzle was better than another way. The tech was often the reason, but speaking techie didn’t help.

While consulting is enjoyable, it’s hard to learn new things when consulting. You spend so much time applying the knowledge you have, it’s hard to have time to absorb more.

And data management changes.

Fast.

If you don’t keep up, you get left behind.

When I got the opportunity for a combination techie and marketing role that would let me learn about the new big data technologies that were coming out around ten years ago, I leapt at the chance. I became a “Technology Evangelist.” Cool title, even if it sounds vaguely religious. And I got to build a Hadoop cluster from bare metal, twice. I got to learn about data analytics, data mining, took a ton of Coursera classes on machine learning. It was glorious. And all I had to do to earn my keep was explain to people some of what I learned. Sweet deal.

I briefly flirted with becoming an industry analyst after that. Did a little work as one behind the scenes, and didn’t hate it. Thought about going in that direction a couple of times. But serendipity had Hortonworks asking me to come document a huge implementation of the technologies I’d studied. That sounded like a lot of fun. So, I was back doing documentation and QA testing, but with the added responsibility of being the communication bridge between a very technical team and some very non-technical executives.

Spent a bunch of time buried to my eyeballs in HiveQL and drawing diagrams so business folks could look at one picture and understand a few thousand lines of code. And technical people could point at the picture to explain what they were doing.

My next adventure was product management. I did not love that. I did enjoy the part of it that had me staying on top of data management technology capabilities, the trends, the customer needs across industries, etc. What I did not enjoy was the cat herding, trying to convince a bunch of other people who didn’t work for me that they should build what I said they should.

I didn’t get to build anything. Or even talk about it outside the company. And when they finally built what I spent a year trying to convince them to build, everyone celebrated the folks who built it, which is legit, but very unsatisfying.

I am always happier when I’m creating.

So, now, I’m back to a technology evangelist kind of role, with an extra dash of talking about how my company’s proprietary software works with the open source software landscape that dominates the data management industry these days.

So, now I get to learn – everything from talking to people cross-industry about their data management challenges and how they meet them to getting certified in new software to taking pretty much any tech training class I want.

I get to create – everything from books to articles and diagrams to new industry terms.

I get to speak – I would have been speaking at conferences all over the world this year, but instead of eating fish and chips in the shadow of Big Ben, I spoke at Big Data London from my home office. Thanks, COVID.

Oh, and yeah, I actually LIKE public speaking, weirdo that I am.

Not much coding, which I miss, but no job is perfect. My boss is totally cool with me doing some coding projects, but I gotta figure out how to squeeze that into my already crazy schedule.

So, there ya go. That’s who is writing this blog.

I have a Medium blog, too, btw, if you’re just not getting enough data management tech info from my sporadic posting here. I also post Vertica specific stuff on the Vertica blog periodically. And I keep threatening to write a technical book on data engineering and architecture in my copious free time.

Keep an eye out. It may happen yet.

Links to upcoming talks in October: (probably should add those)

Oct 20 – 22 Global AI Conference
Python + MPP Database = Large Scale AI/ML Projects in Production Faster

Oct 23 – 25 DataCon LA Unifying Analytics – Drive Decisions with Both BI and Data Science

Oct 26 – 30 ODSC West Unifying Analytics – Combine Strengths of the Data Lake and Data Warehouse


Related Posts