Meet Shreya

Alright – so today we’ve got the honor of introducing you to Shreya. We think you’ll enjoy our conversation, we’ve shared it below.

Shreya, thrilled to have you on the platform as I think our readers can really benefit from your insights and experiences. In particular, we’d love to hear about how you think about burnout, avoiding or overcoming burnout, etc.

I think of burnout as a whack-a-mole challenge. Like an ever jumpy mole, it keeps springing up time and again and honestly, way too soon than I’d like.

I recall severe instances when I was in grad school at Illinois — I had moved from India to the US recently, and I had forced upon myself more coursework than what was required, an internship search, a research assistantship, a part-time job at the university, a position in a university club, and all in the background of a completely new environment, lost connections with family and friends, and an ever-increasing imposter syndrome, since I was among the most brilliant academic company.

I think the first semester is when I had initially felt a strong sinking feeling that I’m stuck and maybe all this (deadlines, submissions, reports) would never end. My solution then was to take everything day by day, never looking past the week lest I become paralyzed with all the pending tasks.

Band-aid solutions like taking a half day break each week or going on vacations every break helped but I never thought of this as a problem that could be resolved. Like every overly-ambitious immigrant student, I actually liked this problem of having more on my plate than I could handle rather than the opposite — not having enough to “be busy”. I realized this much later, and it’s not much of a realization but I really led myself into it. I liked having too much to do, because I absolutely dreaded the other option of not doing enough. Why should I waste my limited time and experience here?

Once the fourth and last semester hit, I started being gentle with myself. I took only a couple of courses that interested me and slowly the rest of the things subsided too — I landed an easier teaching assistantship, had a job lined up in California, and had befriended a few people who could tolerate my jokes.

That helped (surprise surprise). Being gentle and giving myself space really made me happier and a better person to be around. I wasn’t always exhausted or annoyed or difficult to be around. I allowed myself to fail, even in simple things. For example not calling a friend when I said I would, wasn’t the end of the world. Neither was forgetting to read that research paper I was supposed to or not landing that interview at Apple. There is a difference in not being accountable for your actions and allowing yourself to fail.

More recently, I found myself in a similar situation with work and conference deadlines, and moving apartments and upcoming vacations. All at the same time, just as it goes. And one thing I read somewhere changed my perspective: the struggle you’re facing is a dream of many. I remembered how the project I was working on now was what I really wanted to do a year ago. The conference I was stressing about, I was crossing fingers it would work out and I was looking forward to present there three months back. And where I wanted to move, was an amazing location I wouldn’t have thought I would get to live in. And that humbled me; it shifted something in me: I was more grateful than tired to do all this. Sometimes it’s also about the narrative, change the narrative so that you feel humbler yet stronger and more empowered to tackle things.

So I guess tl;dr would be being gentle with yourself and practising gratefulness indeed helped me a lot.

Thanks for sharing that. So, before we get any further into our conversation, can you tell our readers a bit about yourself and what you’re working on?

I work with data and build machine learning models. Currently, I work in the Trust and Safety team at Intuit and build fraud detection models to catch bad actor behavior and misuse of Intuit products. Before this, I worked with the Identity team at Intuit to build anomaly detection and alerting models for key internal metrics. And prior to Intuit, I built and deployed language models for domain name search and recommendation at GoDaddy.

I’ve worked in different domains, but always in data science or machine learning, now more commonly referred to as Artificial Intelligence or AI.

I think the most driving part of my work is to run experiments. It sounds boring, but really it’s thrilling to have a hypothesis and watch it come alive with data! It’s like playing a game with yourself, where only you can be the winner. Because eventually your task is to get the data to predict something and create value. If something doesn’t work, you can twist and turn or prove it doesn’t work and change your hypothesis completely. It’s discovering something new and you get to share it with the world. No one knew that this data just sitting there could predict some actions in the future. You make it happen, you build something new out of the existing. You can be as creative you want with your problem formulation, features, models. How can this not be exciting?

Looking back, what do you think were the three qualities, skills, or areas of knowledge that were most impactful in your journey? What advice do you have for folks who are early in their journey in terms of how they can best develop or improve on these?

These would be my top 3 recommendations:

1. Be a storyteller, and tell your stories to all: Nothing exists in isolation, your work needs to be shared. But the best way to share is to tell people stories. There has to be a plot, a twist, a hurdle, a challenge, and then the ending, sweet or bitter or something that stays with people. And only with practice does this come naturally. It’s easy to get stuck in code or really deep rabbit holes and explain that to data nerds. It’s really hard to come out of it, take five, if not ten, steps back and justify what you were doing there in the first place. The way I try to do this is by presenting in small forums and technical conferences. I’ve presented at a few Python-related conferences and my first talk was so dry and straight, I can hardly listen to it again. But it was a start (a dirty draft is better than nothing) and I’ve come a long way since then. Start somewhere and improve in small but measurable increments. The differences compound.

2. Pick your A-team: Not AI-specific, but cannot stress this one enough. Find your tribe, people you trust and who can pick you out of a ditch-hole at 2 am on a Wednesday. The ones that give you their honest and unfiltered feedback, some tough love when required and who have your back. They could be mentors, colleagues, friends, your parents or partner. But build these relations with trust and lean on them when needed. Learning to ask for help can be difficult, it requires vulnerability, but it should be effortless with this group. For me, my manager and a colleague from my former company come in this circle; I wouldn’t be half the data scientist I am without them.

3. Become excellent in reading data: Any data science role will require you to work with raw data firsthand. What is the data trying to tell you? If you develop an intuitive sense of how to read data, using any tool, the job gets easier. Choice of tool is always secondary: back-of-the-napkin calculations, a phone calculator, an excel spreadsheet, code or chatGPT. Learn how to aggregate observations if the dataset is huge, what kind of statistics and measures will be helpful? What assumptions can you make to help decide faster? What is the basic metric that you can model easily?

All the wisdom you’ve shared today is sincerely appreciated. Before we go, can you tell us about the main challenge you are currently facing?

I’m learning to be comfortable with discomfort. More recently this discomfort has surfaced in the form of uncertainty – both in my work and in general. It feels like things haven’t gone back to “normal” since after COVID19. But maybe things were always like this and it’s only since then that I’ve started being more observant or as I call it, become an adult.

Wars, a plane crash near in my homeland, the political climate around us, uncertainty as an immigrant in my current role, so many factors that could change our lives in a second. I’ve spent a lot of nights panicking and mentally preparing myself for the worst, and there have been periods where I’ve felt I’m in a crisis mode despite being in exactly zero crises.

But what’s helped is again, being grateful, everyday, multiple times a day, and reminding myself that my life has been really fulfilling, so much so that I’d be okay even if this day was the last day of my life; I’ll just have pizza for dinner and call it a night. This sounds dark, but it helps keep things in perspective and strangely enough is pretty calming!

Contact Info:

• Website: https://medium.com/@shreyakhurana11235
• Linkedin: https://www.linkedin.com/in/shreya-khurana/