I’ve had such an exciting time the past year and it has led me into an amazing new opportunity, I’d like to share my learning journey so far!
Modern testing and data science
For the past year, I’ve been inspired by Alan Page and Brent Jensen’s A/B Testing podcast discussions of their “Modern Testing” principles. One facet of Modern Testing (as I understand it) is that as testers, we have to learn how to analyze and learn from production data. How do our users use our products? What errors do they have? How quickly can we act on these? In the era of continuous delivery, we can’t possibly do enough testing to be confident we’re consistently releasing a quality product. The modern tester has to learn about data science.
Continuous Delivery and DevOps
Continuous delivery is another huge area of interest for me. I’ve been working with operations experts since the 1980s, and I’ve always enjoyed configuring CI tools and helping the team shorten our feedback loop for automated tests. I was lucky to pair with Abby Bangser on a pipelines workshop at European Testing Conference this year, i learned so much from her. I really enjoy helping other people learn how they can make their pipelines more robust. I’ll be doing an all-day tutorial with Ashley Hunsberger at CAST 2018 on the Whole Team Approach to Continuous Delivery. I’ll be doing another all day workshop, this time with Christin Wiedemann, “Build Your Own DevOps Roadmap”, at KWSQA. (Am I lucky to co-present with these amazing people, or what?)
Machine learning and test automation
Several months ago, my manager on the Pivotal Tracker team, Jo Webb, suggested that we look into how people are using machine learning for test automation. I’d heard about mabl from Joe Colontonio’s TestTalks podcast. Somehow I also discovered that my friend Stephen Vance works on the mabl team! Just to fast forward this story, I got lots of help from Steve to start doing smoke tests with mabl. I was impressed with how mabl takes in so much information about each UI page and can intelligently attempt to repair a test if it can’t find an element for an assertion or action. Another huge benefit is that each test runs in its own container, so that no matter how many tests you have in your suite, your feedback loop is only as long as your longest test! And, it runs the tests simultaneously in multiple browsers. The speed is really impressive to me.
My husband and I are moving to Vermont, and I want to work remotely from there. Since Steve lives in Boston, I asked him to let me know if he heard of any remote working opportunities there – it is pretty close to Vermont so that seemed like a good option. One thing led to another, and I am thrilled to announce that I will be working for mabl, starting July 30!
We haven’t quite figured out what to call my job, but I’ll be a test consultant of sorts. Once I learn enough about how machine learning can help with difficult test automation challenges, I will start blogging and speaking about it to help others learn. I’m not going to try to sell you on mabl – maybe it is a great tool to add to your toolbox, and maybe it is not a good fit for your context. But I believe that Alan and Brent are correct that we all need to learn data science, find ways we can learn from production use data, find ways we can leverage machine learning to help our teams succeed with continuous delivery.
Let’s learn together! I’d love to hear your own questions about data science and machine learning. I’d also like to hear what challenges your team runs into as you try to succeed with continuous delivery. Many of us deal with “flaky tests”, failing pipelines, lack of visibility of these failures, lots of frustration in reliably delivering value to our customers at a sustainable pace.
Lisa’s Excellent Adventure
I am so excited about this new adventure – moving with my husband and all our animals to beautiful Charlotte, Vermont, and giving myself a huge challenge of learning a new technology. I know I’ll be working with an awesome team, so I’m not too scared. I’m so grateful to the mabl folks for this opportunity.
Oh, and if you know anyone who would enjoy an amazing horse property with a built-in brewery near Parker, Colorado, we have one for sale!