Starting a Career in AI and Giving Back to the Community - Diana Murgulet

We had the pleasure of chatting with Diana Murgulet, a Data Scientist at QuantumBlack. QuantumBlack is an advanced analytics firm operating at the intersection of strategy, technology, and design to improve performance outcomes for organizations. Diana has a background in computer science and machine learning. She is passionate about education, AI for Social Good, and making tech more inclusive.

Topics explored include:

  • Diana's Route Getting Into AI and the Initial Job Hunt
  • Advice for Those Dealing With Imposter Syndrome
  • The Importance of Giving Back to the Community
  • Role Models Within AI and the Value of Having Them
  • A Typical Day at Work Working as a Data Scientist at QuantumBlack
  • How COVID-19 Has Changed the Dynamic of Work
  • The Challenges Faced by Women as AI Practitioners          

Read the full transcript below and listen to the podcast here.

Nikita RE•WORK [0:41]

So it's fantastic to have you join us today for the Women in AI podcast, as well as this video edition that we're hosting. I just wanted to get straight into the first question. Can you share with our listeners and viewers today a bit more about your route getting into AI?

Diana [1:02]

So, I actually have quite a non-traditional route into AI and machine learning. And so, I'm Romanian, I moved back from the UK from Romania for my studies as an undergrad here. And before that, I was not really that much into maths. I mean, I could always do it. And I always had, my parents are both engineers, I've always been encouraged to do the maths, but I was never really very fond of it. So, I guess in many ways, I was really keen to go into humanities. So, when I was about to pick University, it was a really, really hard decision. So, I was very happy when I found this double major degree. So, the UK was just starting these liberal arts degrees, which I guess are more common for U.S. And I started a double major in English Lit and computer science, not exactly knowing which one I'd pick long term, but knowing that I really liked both.

Diana [2:10]

One of the nice things about the Romanian education system is that you do computer science in high school so I'd already done quite a bit of programming, so I knew what I was getting myself into. I then got to uni and realized I really liked the computer science side. And I had great tutors and mentors there who was like, you know, you're good at this, come do this. So, I transferred to the computer science department. And then that was at least the route to tech. The routes to ML and AI was a little bit later when my dissertation supervisor had this really cool project in mind with a start-up in Birmingham, whose main goal was to predict taxi demand in like a kilometer square. And I was like, that sounds like a really cool idea, I'll do it. I was just starting my first ML course. I didn't really know Python, all I had done was Java and C. It's like, we'll try, we'll give it a go. And that's how I basically ended up doing machine learning. I really liked the project. I then ended up doing an internship at Expedia and a master's in the field, and now I'm at QuantumBlack as a Data Scientist.

Nikita RE•WORK [3:29]

It seems to be a common theme at the moment with our podcast is that transition from different backgrounds and different career paths into working in the AI and ML space. So it's fascinating to hear everybody's individual roots. We have quite a lot of listeners, I think, that are starting their career looking into getting into the AI and ML space. So, what was your experience of starting out your career and really with your first rules and your initial job hunt in AI and ML?

Diana [4:01]

It was quite scary to be fair because I had to look at jobs, whilst doing my undergrad. And of course, everyone in the data science community was asking for, at least like an absolute minimum an MSC but most people really wanted people with a Ph.D. and most people wanted people with experience and that was a little bit worrying. I signed up for the masters and I was very keen to do it but then when I was done with my master's I very much realized that no Ph.D. wasn't the way of working that I was going to be happy with. So I started looking for jobs that didn't require a Ph.D. and the list is quite slim or at least it was two years ago, of workplaces that didn't require a Ph.D. or a year of experience. And also, I was going to all sorts of events, I didn't for a second really think that I would end up in consulting, I guess, especially for computer science people, it's not really a traditional path in any sense. Most people want the big four in tech rather than the big four in consulting. So, I ran into someone from QuantumBlack at a networking event. And I was like, that's actually really cool that you solve problems across like multiple industries, and it sounds something that works really well with my restlessness. So I'd like to do that. And that's how I ended up applying, but I think in that sense, it is a little bit hard. I think it's getting better because there are now more degrees in data science, more workplaces are taking data science interns. And I think there are generally more people my age in the field, so it's starting to get better, but it's still a little bit complicated to breakthrough.

Nikita RE•WORK [5:53]

You've kind of touched a bit there on the question I wanted to move on to next, which is something I think you've actually spoken about before, imposter syndrome. So what advice would you give to any of our listeners that may be struggling with that at the moment?

Diana [6:09]

I think that it is super easy to get in tech, especially because there is so much you don't know at every step. And it also relates a little bit to what you were saying earlier with people coming from very different backgrounds into the AI space. So, everyone will have a very different experience, and like I said with super-specialized knowledge. So, I think what makes it scary is that technically everyone is an expert in their own little slice, and they will be better at it than you right. I think for me, it was a matter of, first of all, just growing a bit more confident in my own skin and learning to ask for help. But I think the general piece of advice there would just be to find an area that you're really, really excited about because then the learning comes easy. And you find it so easy to read more into it, to go to conferences, to talk to people about it, and then you do become the one specialist in this small thing. And then the imposter syndrome slowly disappears.

Diana [7:21]

I think the other thing that we've been talking about in an article on Medium that I authored with a colleague was the whole, do one teach one, sorry, I'm going to get this wrong. See one, do one, teach one idea. The fact that you are going to have to learn from those around you and you're surrounded in the workplace by amazing gifted people that are willing to teach you. So you should observe and absorb from them and then you should also practice on your own , grow the skill, and then it becomes your job to teach one. So that can be speaking at events, that can be just helping out a more junior colleague, but it then gives you that confidence that you've actually mastered this to the level of helping someone else with it. So I think that that's definitely helped me but it's also to some extent, just a matter of time.

Nikita RE•WORK [8:20]

I think that's really good advice for any field, you only feel that you really have mastered something when you can sort of teach somebody else to that same level. And you also touched on there a bit about asking for help, if kind of when needed to help you progress. That's something that we definitely speak a lot about on these different podcasts is how important it is to give back to the community especially mentoring women that may be at the beginning of their career. What kind of value do you see in that from the mentorship site?

Diana [8:56]

So I absolutely love this and I really love that I work for a place that has this so ingrained in its identity and its DNA, so QuantumBlack and McKenzie do a lot of work both in the space of AI for social good but also working with a lot of incredible organizations that they help and support. So we do, at least on the tech side, we work with teams in AI, we work with Code First Girls, we work with Career Ready, with a lot of other organizations where this is the key of what we do is take this skills and help other people grow. I've always been very happy that the company I ended up working for, gives me these opportunities, and already had these great connections so I can do it inside the workplace. And there's, we're split into guilds in QuantumBlack, and we have the data scientists and the data engineers and the software developers and the designers and each of these corners kind of play their role to help people in the field. So I think that's the one thing that makes me happy in terms of that. Then I guess there's the more informal side, I'll always try to make sure that if people reach out to me on LinkedIn or anywhere else, friends from uni or colleagues or people I know from other circles, just make sure that I can give a hand especially when people are trying to follow in the same footsteps or get into tech or get into the same companies. I am very happy to just speak about the experience because I think one of the things is that you don't know the workplace you're going to end up in until you're actually there.

Diana [10:44]

So the more you speak to people that are in these companies, the more you understand the culture and the values and all of these things. I think that's, at least in terms of, what I get to do as part of work. But then there's also the component of things I get to do in my personal time. I've always really, really enjoyed volunteering and giving back in that time. So I'm now working with a Romanian NGO called Casa Buna, which gives private tuition to kids from disadvantaged backgrounds. So once or twice a week, I have a mentee that's 11. And I work with her on like maths, and I work with her on her reading skills. And I guess it's a very different type of experience because it's very much not tech and it brings me back to my love for teaching and what we were discussing earlier, right. But it's such a nice experience to be able to shift your focus entirely from this corporate world that you have your full attention to, back into the real world and how things operate. So I really enjoy that.

Nikita RE•WORK [11:58]

You'll have to share with me the name of that organization and we can add it to the notes at the end of the podcast for anybody that is interested. And for yourself, how important have role models been for your career, and kind of for your individual path?

Diana [12:16]

Very, very important. And I think that the definition here of what makes a role model is very broad from the people you look up to in a public setting of like, the amazing women that I see climb up career ladders we've just had the sad passing of Ruth Bader Ginsburg, who was an incredible icon in that sense of what you can do and what boundaries you can push as a woman. But then there's so many tiers and so many layers where you can look up for role models. I think of the women I work with, and seeing them in action and change things makes me make me very happy. And then you have the teachers and the lecturers we were discussing earlier, the people that encouraged me to pursue tech. That was very valuable to me and in both the mentorship and the looking up capacity. I think the workplace, mentorship, and role models setting is also very important and very valuable. The McKinsey London office was led, until recently, by an amazing woman named Vivian Hunt, we were just watching her this morning in a chat with Berendina Veresto. So that was very cool to just see two very powerful women who have achieved incredible things, one in terms of literature, one in terms of the corporate career ladder. So it is very nice to be in a workplace that offers you that and to also be able to look at the world a bit broadly and say, oh, look like things are changing, both in the field and outside of my field. And yet also in personal life, in terms of friends and everything.

Nikita RE•WORK [14:18]

Definitely, I think it's important for any stage of your career is having those types of role models to be inspired by throughout your whole career path. So in terms of, you mentioned a bit about McKinsey and QuantumBlack there, but what does a typical working day look like for you if there is such a thing these days? And what industries and topics are you currently focusing on that you could maybe share a bit more about?

Diana [14:46]

So I think what I was saying earlier is that our company's quite interdisciplinary, both in terms of like the clients we serve, the industries we work in, but also in terms of the people that we have. So our teams are very mixed, which means makes the working environment very fun because you get this knowledge exchange and these learning opportunities all the time. So in terms of a day, I'll normally be working on a client project with one of the teams and as I was saying, we will have a data engineer and a designer and you know, an engagement lead. So most mornings, we'll just start with like a standard check-in and discussing what the plan for the day is, then I'll probably have a couple of hours of meetings which, unfortunately, these days will be over zoom. Which is where the collaborative element of it is very important. I really miss just whiteboard exercises and just being able to have that exchange of ideas happens a bit more naturally. And then one of the important things in my day is saving some time for deep work, so for coding, and actually getting to think about the problems that I have to solve. And then depending on the week, we're talking about all of these volunteering opportunities, there'll be all sorts of those included in the schedule. So it's not always the same, but that's the structure.

Nikita RE•WORK [16:14]

And we're recording this on zoom today as it is quite common, you've just mentioned that's being a part of your meetings too. And how else has COVID-19 changed that dynamic of work for you, over the past few months?

Diana [16:26]

It's been rough in a lot of ways. Because I really, I really, really like people and I really like working with people. And the nature of our work is very dynamic, very interactive, and it relies so much on building relationships, both inside the team and with our clients. So it was very hard initially, especially the way consulting works as you'll move from project to project and work with new people. So in a lot of these cases, you'll be working with people you've never met before, from other offices. So you have to create now this bond and this working relationship just over zoom without having had any contact with the person before. I really miss the in-person interactions, I really miss the ease of just moving to a whiteboard and someone says this, and we throw another idea in. We found ways to work around it, but it still feels like it's not the same.

Diana [17:28]

I think the other the other thing that I really miss is the small interactions, like being able to run into people in the kitchen and just discuss stuff like unrelated anything for like 10 to 15 minutes. And the types of connections that are formed through that and the bonds that strengthen, now require far more effort, you need to actually take active steps. We have a really cool thing that we've set up, we had it before COVID, but it's played an even more important role now. We have what we call random coffees, where like people in this list get assigned a random partner and they have to meet in the next few weeks and just have half an hour coffee, over zoom, which puts you in contact with people that you may have not spoken to before. And then we set up this afternoon tea that also would have happened organically in the kitchen on a Wednesday but is now a zoom room where we can all come in and catch up. And it's not work-related and it gives me this like breath of fresh air to recharge us a little bit.

Nikita RE•WORK [18:37]

It is so important to continue to do things like that. I know there's been a lot of talk about zoom fatigue and there are definitely drawbacks to us having to communicate over that. But then again, we're very lucky to be in a situation that we can continue to do the work that we do whilst we're still working from home, so it could be worse. And just finally, I wanted to touch back on to the woman in AI aspect of this podcast. So what are your thoughts on the challenges faced by women as AI practitioners?

Diana [19:12]

So we were discussing earlier, I'm quite hopeful about this, but without being too optimistic. I think I tend to lean on to the side of realism here. I do think things are getting better, and the whole women in the workplace debate is getting far healthier. We're becoming more used to reporting things when things go wrong, and observing inequality, not just in terms of gender, but there's been a lot of talk about race recently. So I think things are getting better. There is, however, an element of, that has been reported constantly the last few months, that this crisis is going to push women back. Women's workloads have increased significantly since the start of the crisis and we're at a huge risk of just moving back overall, like leave alone the AI label for a second. I think in terms of women in AI, I found it incredible that there's a super-strong community there. Those groups of women in machine learning and so on, do a lot of work to create a platform and an environment that fosters this culture of creativity and encourages women to feel an integral part of the field.

Diana [20:47]

So I think overall it's getting better, I think we're still fighting with a lot of imposter syndrome. I think there's still a battle there around uncomfortable remarks that you'll hear at some point or the other. But I think people are becoming more, I think the fundamental shift has to happen, like far earlier than getting to a career in AI, I think we're putting way too few girls through the tech pipelines. I think we're encouraging way too few women to go into STEM to start with. So I think the moment we fix those things further down the line at the beginning of the pipeline, we'll definitely see more women in. I think, for instance, at my level at the start of my career two years in, I definitely see that there are too few women in senior positions. I can look in senior positions in other fields and in other disciplines, but in terms of tech, there are definitely too few women. And we also know that a lot of these women drop in their mid-30s and make that family choice there. So I think once we fix the pipeline overall, and we can start seeing some of these role models far more often, then things will get a little bit better.

Nikita RE•WORK [22:08]

I hope so. So many important points there right from connecting girls in STEM, right through to specifically working within the AI space. But you're completely right in that it's not specifically an AI issue. It's a challenge that is seen across all of the technology, and not just technology as well, but women in business. So we've actually seen in the past few months, whilst we've all been changing and impacted by COVID-19, we've seen a lot of enthusiasm for working with women in AI and so that's, I think there's a real positive to come from that. And lots of different companies and individuals have been discussing with us over the past few months about what we can do in the space. So I think there are lots of potentials there and we're going to see lots of good things going forward. Thank you so much, Diana, for taking the time out of your day today and for joining us over zoom, and for having a chat about some of these really important issues. It's fantastic to have you involved and for any of our listeners that are keen to connect with you, what would be the best way to do that? Would it be on LinkedIn or Twitter?

Diana [23:21]

LinkedIn.

Nikita [23:22]

Okay, fantastic. Well, thank you again, thanks so much for taking the time to chat with us today. And we'll speak to you soon.

Diana [23:29]

Bye, thank you.

Nikita RE•WORK [23:36]

A huge thank you again to Diana for taking the time out of her day to chat with us this week. It was fantastic to learn more about her route getting into AI as well as the challenges of imposter syndrome, and also giving back to the local community. Our Women in AI Virtual Evening is only one month away, so if you're interested in getting involved and connecting with more women in the field, then check out our website re-work.co/events to book your place. Take care, until next time.

Are you interested in reading more AI content from RE•WORK and our community of AI experts? See more blogs below: