My Journey to Data Science – Part 3 of 3

My Career Journey Over the Last 20+ Years

October 28, 2016

 

Jamey Johnston (@STATCowboy)

 

This is Part 3 (the final post!) of a 3-part Blog series on my Career Journey the Last 20+ years into my current role as a Data Scientist. Part 1 can be found here and Part 2 can be found here!

BI

 

BI is fun! I loved it and soon was not missing the late-night pages of down databases or late nights doing database upgrades. (Although, one night I upgraded 11 Oracle production databases to Oracle 11g at the same time which I must admit was fun!) I loved the insights I gained from building BI solutions using MS products like SSAS, SSRS, MSSQL and SharePoint. I also loved the new-found freedom of exploring data and making it all work together in tools like Spotfire and Power BI! I was building large analytic datasets with billions of rows and special analytic calculations. I was having fun but I was missing something to truly fill the gap as a data scientist, statistics!

 

TAMU MS of Analytics

 

I was driving late one night into the office on I-45 in Houston which was not my normal route and I see this billboard for Texas A&M Masters of Science in Analytics! I had been looking for a program to learn statistics but I wanted it local (my ADHD doesn’t do well with online classes!). This was March of 2014. I was accepted into the program in June 2014 and started classes the end of August 2014! To say the least, the next two years my life was crazy! I had started to learn R the last year or so and in the program, I learned about SAS and SAS JMP and a ton of statistics from regression, time series, spatial statistics, multivariate analysis to machine learning! I was having a blast learning statistics and all my friends and family thought I was plum mad to start grad school in my 40s and in all things, Statistics!

I can’t emphasize the importance of a good education in statistics and how valuable the MS in Analytics from TAMU was in that education. I graduated in May 2016 with my Masters and could not be happier with my experience at TAMU. I am proud to be a member of the Fightin’ Texas Aggie Class of 2016! Yes, I wore Mickey Ears with my Twitter handle (@STATCowboy) on the back at graduation!

 


 

Data Scientist

 

Armed with my learning of R, SAS and more importantly, Statistics, I am fully ready for whatever data is thrown at me. I get to build some really cool statistical and machine learning models these days and I can’t believe at 42 years of age I have a new career I basically started at 40! I guess if you are going to have a midlife crisis a career change to The Sexiest Job of the 21st Century according to the Harvard Business Review is a good crisis! Who knows what career I will have at 50?!

I’ll finish with my definition of a Data Scientist – someone who loves finding insight from data, big or small! In some ways, we are all data scientist and I would encourage anyone who would like to take their data knowledge to the next level to learn Statistics. Statistics is the first leg of a three-legged stool that makes up data science. The second leg is a data guru! Someone who can store, move and manage data and present it in an easy to use fashion that can be consumed quickly. The final and third leg of the stool is a Subject Matter Expert (SME) who knows the business and the data around the business. One person is never an expert in all of these but a Data Scientist is typically an expert in one and very proficient in the others. However, I find the best Data Scientist is really a team of folks that love data and bring expertise in one of the three areas – statistics, data or subject matter expert – to the table to build models to make business better!

To get you started in your career as a Data Scientist check out my presentation from SQL Summit 2016 here. Be sure to check out the resources section of the PPT to see some good locations to learn things like R! Also, feel free to reach out to me on Twitter if you have questions along your journey. I am happy to help!

I hope you enjoyed reading about my journey. I know have enjoyed living it!

My Journey to Data Science – Part 2 of 3

My Career Journey Over the Last 20+ Years

October 26, 2016

 

Jamey Johnston (@STATCowboy)

 

 

This is Part 2 of a 3-part Blog series on my Career Journey the Last 20+ years into my current role as a Data Scientist. Part 1 can be found here!

DBA

 

I had been working for LDEQ for three years now and was one year removed from college when a good friend called. He was working as a consultant for Oracle Corporation in Houston. Oracle had signed a big contract with a client in Houston and needed DBAs on the project. He told me to send my resume so I could join him which I did! Two weeks later I was a Senior Consultant with Oracle Corporation working and living in Houston. It was a very exciting time of my life those three years. I had one of the best DBA mentors you could have, Clement C., teaching me about Oracle and technology and was getting to travel to some cool places in the US. Most importantly, though, I found the love of my life soon after moving to Houston, my wife, Melanie!

Oracle was a great 3 years for me and I learned so much that truly set the stage for the next 10 years or so of my life as a DBA. I enjoyed Oracle but after 3 years of travelling it was time to stop traveling and get a local job in Houston so I could spend more time with my wife and dogs! My wife will tell you I only stopped traveling for the dogs (not true though)!

This little-known company at the time called Enron (Ha Ha!) was hiring and I got hired on there! It was June of 2001 and Enron couldn’t be flying any higher (at least in their make-believe world!). We were one of the 10 largest companies (at least on invisible paper) in the world and the talent coming to work there was second to none. Over the next year and half I would learn a great deal about business as well as management as I was soon promoted to manage one of the core DBA teams at Enron. Enron went bankrupt and I moved to work for UBS when they bought the North American trading division of Enron.

After a brief stent at UBS they decided to get out of Houston and I went to work for a small consulting firm, Methods Technology Solutions, out of Baton Rouge. I did some travelling again and really had a lot of freedom to learn some new things and work on a variety of database projects from very small projects to bigger ones. I worked for Methods for 4 years, the last year and half consulting at a large independent Oil and Gas company. The Oil and Gas company offered me a full-time employment job and I have been there now for over 12 years.

The first 7 years at the O&G company I was a DBA team member working on admin, upgrades and a lot of tuning. It was fun! I had some time to write a little code, too! I created an end-user web dashboard for the app owners to see what was happening in the DB with their systems in C# ASP.NET. I created a multi-threaded application in C# to rapidly put together archived emails in to a bundle and create audit trails in XML, CSV, TXT and DB tables. However, I mostly did the typical enterprise DBA tasks.

 

Tuning SQL to Enhancing Business Processes and Relationships

 

I really enjoyed tuning database systems. I was a master of 10046 traces in Oracle and SQL Profiler and Extended Events in SQL Server and was able to tune a great deal of our applications. The funny thing about tuning databases when you are a DBA is you tend to look for that magic parameter, trace flag, bad SQL, etc. and often forget about the application and business processes behind what you are trying to optimize. More importantly, you often forget about the business people using those applications and how that data is used to add business value.

Anyone who knows me knows I am not shy! I love people and I love engaging with people. This led me to have a different approach to tuning. When I would get a call that something was slow I would find the business person who was having an issue, schedule an hour of their time in their office, have them reproduce the issue in the system while I watch both what they were doing in the application and also what was happening in the database. It was funny to see their face when the “DBA” would show up in their office. It was like an endangered species had left their native habitat and ended up in their office. Many would even tell me I have never seen a DBA before! This approach created a special relationship that allowed me to start understanding the business processes behind my company and fostering relationships to do more than just “tune database systems”.

I was put on a special project to implement a new business system and was actually placed for a year within the business team to assist with implementing the application. The application is a critical one for O&G companies and so I was dedicated to just working on that project and application. For the first time in almost 15 years I wasn’t a DBA! I wasn’t on-call (I didn’t know what to do with my new-found freedom). I didn’t have to do database upgrades. It was so weird to actually be a “customer” of the DBA team! It certainly gave me insight into how it felt to be on the other side of the table. I absolutely loved the project. I spent so much time understanding all the various data pieces and how they would flow into the new system (we had several different DBs that would need to integrate into this new system) and more importantly how that data truly worked for this particular part of the business. I really loved data! I realized then that I didn’t really want to be a DBA full-time (still like the part-time idea just to keep current but not very practical) but I wanted to work with data. I wanted to build datasets and analyze them. I wanted to find ways to better connect them together. I wasn’t quite on the statistics track yet but definitely the data and business side of Data Science.

 

DBA to BI

 

As I began to understand the applications and business processes behind using the databases I supported, I also became aware of how so many of the systems were connected to one another either by ETL, database links, linked servers or virtually through various business processes. The connections of all the systems and more importantly, for me, the data led me on my next journey into BI.

I had become exposed to a tool called Spotfire while I was on the special project and learned how to use the tool and how people at work were using the tool. So, I began to work on connecting Spotfire directly and securely to the data sources and soon I was moved onto the BI team working on BI projects in Microsoft and Spotfire. I was a BI guy now!

 

Stayed tuned for Part 3 tomorrow – BI to Data Scientist!

 

My Journey to Data Science – Part 1 of 3

My Career Journey Over the Last 20+ Years

October 25, 2016

 

Jamey Johnston (@STATCowboy)

 

So I am at the EMP in Seattle attending the SQL Summit 2015 Appreciation Event and I am introduced to Denise McInerny (t) by my good friend Wendy Pastrick (t | b). Of course, the common pleasantries are exchanged, “My name is …”, “I work at XYZ company …” and “I do ABC job …”, which in my case is “My name is Jamey Johnston, I work at an O&G company and I am a Data Scientist”. Denise’s response was not quite what most people’s response is when I tell them I am a Data Scientist. Usually I get a general look of trepidation or the occasional, “Oh, you are an unicorn!” (true story, several times), but in Denise’s case she says “You should write a blog about your journey to become a Data Scientist” (or something along those lines). I thought that might be a fun blog to write and said “Sure!”. So here is the story of my journey to becoming a Data Scientist.

 

10 PRINT “My Name is Jamey”

 

So, I am an eight-year-old living in the suburbs of New Orleans and I want a computer, a Commodore Vic-20, and I get it! I plug it in and connect it to the color console TV in my living room and turn it on. Within 10 minutes I have written my first program:

 

10 PRINT “My Name is Jamey”

20 GOTO 10

30 RUN
 

The next thing I see is the greatest thing I have ever seen – “My Name is Jamey” scrolling across my parent’s TV screen as fast as I can read it! I sit there and watch it go across the screen over and over and over again finally asking myself – “How do I stop it!?”. I scan through the manual that came with the Vic-20 looking for the correct key combination to stop it! No Luck and no Internet in 1982 so I do the only thing I can think of to stop it – “Unplug!”. I loved that computer and it was what fueled my desire to learn more about computers!

 

GIS and College

 

Fast forward to college and I am a 19-year-old who just finished his first year of college as a Psych major and I get a job with the Civil Engineering department at school working with GIS and Remote Sensing technologies. My father had started in GIS and Remote Sensing back in the 70’s for Wetlands research working for the Department of Interior so I was familiar with GIS and Remote Sensing and I was excited to work with big workstations, lots of data and getting to work in the same field as my dad! So these big workstations were SGI and DG boxes running from 25Mhz to 100Mhz processors! Your phone is probably 100 times faster!

Two years later I finish my junior year and I am still working on GIS and Remote Sensing projects, one of which was to run clusters against tiles of Thematic Mapper satellite data of the entire state of Louisiana and then use those clusters to classify them into land use and land cover categories (i.e. is this cluster water or agriculture or coniferous forest, etc.). I was working with ESRI and Erdas Imagine software and learning UNIX which was really fun and beneficial to my career.

I loved being a GIS and Remote Sensing technologist but I was still a Psych major! I wanted to change majors but the problem was my University only had ONE geography class! So there was no way I could stay at the University of Southwestern Louisiana and get a GIS degree. I transferred to LSU to start in the Fall of 1995 which had a great GIS program through the Department of Geography and Anthropology to get a BS in Geography with an emphasis in GIS and Remote Sensing. I also went to work for the GIS Center at the Louisiana Department of Environmental Quality (LDEQ) continuing my career in GIS and Remote Sensing. It took me two more years to finish school as I had to take 36 hours of Geography to graduate. My first semester was 12 hours of Geography with a Geography 1001 class up to a Graduate level class of Geography 4998 and a music class I believe. It was a fun semester to say the least.

I graduated in Spring of 1997 from LSU and continued working for the GIS Center in the field of GIS and Remote Sensing. I also started working with another cool technology, Relational Database Management System (RDBMS). ESRI had created the first version of ArcSDE which was a way to store your spatial data in a database. In my case it was Oracle. We had Oracle v6 and v7 databases at LDEQ and I was starting to learn them to support ArcSDE along with some other technologies that required RDBMS.

 

Stayed tuned for Part 2 tomorrow – DBA to BI!