This week we welcome Maria Khalusova ( @mariakhalusova) as our PyDev of the Week! Maria works for JetBrains and will be speaking at AnacondaCON this April. If you’d like to catch up with her, you can check out Maria’s blog. Let’s take a few moments to get to know Maria better!
Can you tell us a little about yourself (hobbies, education, etc):
Growing up I loved two things – math and books. I kid you not, I solved math problems for fun. Not surprisingly, I went on to study Applied Informatics at the Dept. of Mathematics and Mechanics of Saint Petersburg State University which I graduated from in 2007. This field is actually really close to modern Data Science, all the math parts of it were there, plus a good chunk of computer science program. I only wish I got to learn Python at my University and not Java 😀 Sadly, back then such fundamental packages like pandas and scikit-learn didn’t even exist yet.
Even before I graduated, I started working at JetBrains, first as a technical writer for IntelliJ IDEA. Fun fact: this June will be 13 years since I joined the company. I’ve changed projects, job roles, countries even, but not the company.
In recent years I’ve re-discovered my passion for math, data science, machine learning, and deep learning. I’ve brushed up on my rusty math knowledge and have been self-educating ever since. This is a never-ending process which I think is great.
Why did you start using Python?
Once I realized I wanted to do data science and machine learning, the choice of the language was quite obvious to me. There are of course R and Julia, but the two major things that got me sold on Python were the deep learning libraries and Python being a general purpose language.
What other programming languages do you know and which is your favorite?
I mentioned earlier that back in the university I learned Java but it never fully clicked with me due to all the boilerplate one had (still has?) to write to do even a simple thing. I do know some R but I don’t really use it in anger. These days I’m eyeing Kotlin but haven’t really used it yet for anything other than toy samples. Shame on me, given how close I am to the people who are the ultimate experts on Kotlin.
What projects are you working on now?
At the moment of this interview, my main focus is working on a talk I’ll be giving at AnacondaCON this April. The talk will be about machine learning model evaluation metrics, what they are and when to choose which one. This is something that caused some confusion for me as a self-educated machine learner, and I want to help others get a better intuition of these essential metrics.
Another big thing for me is organizing PyData Montreal meetups. I’m new to this and this is both exciting and a bit terrifying 😀 Our first PyData Montreal meetup exceeded our expectations on all fronts! We’ve had about 70 people show up and the audience was really engaged with the speaker. Christian S. Perone has delivered a concise, interesting and practical talk, “PyTorch Under The Hood”. I feel overwhelmed by all the positive feedback we’ve got and can’t wait to have our next meetup!
At work, my main responsibility as a Developer Advocate is to serve as a communications bridge between the community and development teams: collect feedback, learn about trends, advocate for community’s needs inside JetBrains and help our users become badass at what they do. I’m also helping our marketing team and content creators with analytical tasks.
In my spare time, when I have it, I dabble with Kaggle competitions, I scrape data for pet machine learning projects, and recently I got myself a raspberry pi, so now I just need to come up with some fun IoT idea 🙂
What non-Python open source projects do you enjoy using?
Non-Python? Not that many, really, come to mind. I find reveal.js great for generating HTML presentations, and I’m using Jekyll for my GitHub-based blog. In my smaller pet projects, I often use DB Browser for SQLite. I’m occasionally using Apache Zeppelin.
How did you get into doing conference speaking?
I visited a lot of conferences as an attendee, and at some point, I had an idea that I actually had something to share myself. I submitted a CFP, got accepted, completely freaked out, then prepared and delivered the talk. To my surprise, people not only showed up and didn’t leave, but they generally liked it. I loved the experience itself and the conversations I had after that. I simply got hooked.
What kind of advice do you have for others who want to speak at conferences?
Hadi Hariri recently wrote a series of posts on public speaking with a ton of helpful advice, I would genuinely recommend reading all of those seven articles, they are gold. Here’s the first one.
What is your motivation for doing talks?
Imagine teaching a friend something that you just learned or maybe known and practiced for years, but something that you’re excited about, something valuable, something that you think will help that friend with whatever she’s doing or open up new possibilities.
Now, giving a talk scales that. Public speaking, of course, requires a lot more prep work than talking to a friend. You need to do your homework, research various angles of the topic, make sure your content is coherent and not boring, your message is clear, the timing is right, and so many other things. But! As you do prepare, you learn something new along the way every time – maybe a different point of view on your topic, alternative technology and its pros and cons, related scientific research, tricks for making better presentations, or something else. You also get to educate and inspire folks in the community, spark conversations and generate new ideas from them. To me, these are great reasons to speak at events.
Is there anything else you’d like to say?
We’re looking for speakers for our PyData Montreal meetups, so if you do Data Science, ML or DL, and want to share your knowledge or experience with local folks, please feel free to reach out on twitter: @mariakhalusova or @PyDataMTL.
Thanks for doing the interview, Maria!