DoorDash Data Science Interview Process 2026 (I Got My Highest Offer Ever)

0:00 I got rejected from Door Dash in May and then fast forward it's October now eight interviews later and I'm going to be joining Door Dash as a senior data scientist and in this video I'm going to share with you entire process. I'm going to share with you how I prepared for the interviews and what they ask in interviews and basically I'm just going to share details about my future job and my role and what I'm going to do in Door Dash. So, if you don't know me, hi, my name is Andre Yinski and on this channel I share career and life advice. First, in Maine, I applied to Door Dash. I called apply without a referral, cover letter or anything, and it just got rejected straight away. Reasons for rejection when you apply that way can be various. Maybe the role got filled, maybe they didn't like you for some reason, maybe just they used AI for your application and AI didn't like you. So there are so many reasons why they didn't select you in the first place.

0:57 But if you invest in the LinkedIn and if you do it the right way, if you increase your profile visibility and you network on LinkedIn, then you have chances for recruiters and managers to notice you directly. And when you do that, there is much higher chance that you're going to pass the screening phase because applying like it's the hardest part. Even if you're a senior and especially if you're a junior, it's it's just hell. It's hell out there. If you're just starting or graduating, it's going to be so hard to cold apply. You need to find people and establish connections directly. And even if you are senior, it's still hard. So manager from a data science team in Door Dash reached out to me and said I like your background.

1:42 Let's have a call and I'm like all right let's see what he got because I just got rejected from them in May and then it was happening in July. We had a call, we had a great chat. So basically what happens on this first call is just they evaluate whether or not you are like in the ballpark of being the right person, whether you did what they need to do, if you can communicate or not. They just like to put a face to your name and it's not hard at all. So we talked for about an hour and I was like okay I want you to come and complete all of these interviews and remember I had eight of them in the end. So it's just it's a long process and it all started in July and my first day is going to be in October. So it's it's just crazy. So the first one is technical screening. It used to be that you would code in SQL and solve a business case during a technical screen. My process was that I just had a case and it was a standard case for a data scientist. I cannot give you specifics because I signed an NDA but it's it's going to be something like this. We have some kind of problem on our platform or we add in some feature.

2:48 Could you tell me about the steps that you would go to diagnose this problem and define success metrics for the feature and then like set up an experiment for it and what would be the problems? What would be the difficulties? Of course, you need to know all of your basic data science stuff. Of course, you need to choose right metrics. If you're not sure which metrics to choose, you can always read companies analytics blog like Metahazard blog and Door Dash has a great analytic blog and it should be your main resource for preparation. But beyond that, what you need to know is when you interview at Door Dash, it's a threesided marketplace. So you have users on one side that order food, right? And then you have merchants on the other side, restaurants preparing the food. And then you have dashers and cers that deliver food. Like imagine you added a promotion to some area like 50% off on everything and then you have people ordering more in that area and that exhausts all the dashers resources, right? So because of all the people get orders in one area and then all the orders go in the same area and then of course you have problem with other people. So you need to always take that into account and that means that usual methods of casual inference like experiments are not always usable.

4:09 So you have to work around with that either with switchback experiment or with some other methods like differences in differences or propensity score matching. You always need to make sure that you account for it. So after the technical screen came good news and bad news. They said, "Good news, Andre. You are great. We want you to come to on-site, which is another four interviews, but the team that you initially interviewed on, they filled the position." And I'm like, "Okay, and here's when you work with a recruiter."

4:42 I could have said just, "Oh, okay. Well, that's too bad. Bye-bye." But I worked with a recruiter and I had a call with them and I said, "Well, find me another team, right? It's only logical. I I might be suitable for another team because the manager liked me and yada yada and then they did and I said okay we have another team for you interview for them. During on-site what you have is four interviews. What's different about on-site is of course the bar is higher like you know when you pass technical screening like approximately a quarter 25% of people who got through the first resume screening pass technical screening and then about 20% of them pass onsite. So like it's a tough selection tough funnel but apart from that people who interview you for your onsite are the people that you're going to be working with. So for me it was data science managers and directors of their organization within Door Dash that I would be a part of. So not only they look at how well you solve case or not they look at the same time whether or not they want to work with you. So interviews first one is SQL technical challenge nothing too fancy if you know group bytes where joins window functions you're going to be fine. One exercise was particularly hard because it was you were given a piece of code that you didn't write. Somebody wrote it and you had to tell what it does. It's always so difficult going through somebody else's code and then you had to change it so it does a different thing completely. So it was novel. I didn't I didn't really see the challenge before. But if you like go to data leour or interview query or any other platform, you're going to be just fine. like it's if you use it every day, you're going to be fine. And then three other interviews were similar cases as technical screening but more intense and we went deeper into some of the methods.

6:44 It was mostly about experimentation and data. a little bit was about ML models and how to evaluate them but mostly it was how we can choose the right metrics, how we can use data to come to a decision. What kind of data would you use? and then various problems that they had during their experiments. So it was very you know back and forth. It wasn't like somebody was lecturing me. It was like I was sitting with a colleague and we would brainstorm some situation. What is important here and I learned it with my meta interview and I had it twice I failed it twice the on-site stage. The important thing is that you communicate all of your thoughts and that your thoughts are structured. So the problem that I had during my meta interview is I would get a question and then I would start answering and then at some point I understood that I didn't really get the question. So first you need to like really make sure that you know what what's being asked and you know the goal and you know what they want to hear. And then the next part my problem was I would be speaking too long and I would add details and details and details during my meta interview to my answer and then interview didn't really have a space to kind of provide feedback at the same time. So you need to have a clear start and end of your answer and then you can clarify and interviewer knows okay he didn't mention this or he didn't mention that. Let me guide him a little bit closer to the point. If your answer looks like never ending you know spree of words you say we can choose this success and this metric and this metric and that and also that and oh you know what I forgot about this and if your answer is anything like that and you take too long then you're just out of timing for interviewer input. So you need to provide short concise answers and that what matters and you always remember like this is a dialogue you you're speaking with a live person they're not a professor who are testing you they are your future colleagues so ask them questions challenge them and always remember always provide reasoning for your answer in cases like this when you have a product case or a business case or a data science case usually it it doesn't really matter what you answer. What matters is your reasoning behind it. When you're choosing like a metric, do you understand the trade-offs of the metric? Do you understand the limitations of the metric? They looking for people who are conscious and very very specific about what they do. So you're mentioning like a set of metrics and you always go through tradeoffs and problems and limitations and reasoning why you choose what you choose. So always tie it to the reason and always tie it to company's goal. Once you pass that and like in about a week or two I got a message like you're past that.

9:56 Hooray we're in the final stage and final stage is behavioral interview. And problem with behavioral interview is many people underestimate how important it is. Let me tell you something. A lot of big tech companies, they not only evaluate whether or not you're going to be hired by the company by behavioral interview, they evaluate on which level you're going to be hired and what amount of compensation you're going to getting based on which qualities and signals you provided during your behavioral interview. A lot of technical professionals developers, engineers, data scientists, they just say okay I can code this doesn't matter. It matters a lot. So you need to prepare. The way you prepare for behavioral interview for the companies is you go to the company website, you learn everything that they do, their context and most importantly you learn what they care about, their values, culture and mission. most likely you're going to be evaluated against those values. So when you prepare your stories like they're going to ask you tell me about a conflict that you had recently you need to understand that you need to signal that you possess those values and qualities that they're looking for. So you transform like qualities that they have like we want we move fast or we are oriented on impact or we are flexible. you transform that to kind of your actions and verbs that you demonstrate through stories. If they want to see somebody who's agile or somebody who's impact oriented, you got to show it through your story. Like you got to give in your story when answering the question several like kind of signals indicators that you have those qualities. I do career consulting and I help other people and I shadow their interviews and I listen to what people are answering. Sometimes people don't really understand what's being asked and like you can get into this trap when you ask the direct question but rather I want you to think what really they want to know when they ask this. I I'll give you an example like they're going to ask what are some of your weaknesses? Tell me about a recent feedback that you received. And usually a company doesn't want to hear that you are anxious or disorganized or this or that. It is a story. They want to hear how well you work on your shortcomings. So basically a guideline for this story would be I searched for feedback from my manager and I got this feedback. I worked on that and here's what happened. Here's my growth story. here's how I overcame this weakness. usually they want to hear that you are capable of change when they ask this kind of question. And you see how a simple question if you just answer it at face value it's it's not going to be good enough. Right? You see how you need to think what they really ask when they ask this question. And nobody's going to tell you this but this is what they look for. And the same is with other behavioral questions. So, I just want you to double down on preparation and I just want you to be very conscious about what you're answering them. Then after that, they said, "Okay, we're ready to make you an offer." Like, you have a great scores on your on-site on behavioral. Usually, what happens then is they get your package, everything together, and they calibrate and they talk whether or not they want to hire me or not. And they arrived that we want to get this person. and I had an opportunity to have a call with my manager and with the director of an orc that I'm going to be joining in. It's a very very unique work. It's working directly with merchants and restaurants.

13:48 Basically, what is doing is is improving integration of Door Dash and helping restaurants grow their value and revenue. And it's very very exciting because it's not a core product and it's something that practically new and novel and it's kind of like a startup within a door dash. They're trying to find ways and new ways how they can optimize restaurant experience and it's just something that was very exciting for me because entirety of my career I worked in startups and like when they said it's like a startup within a door dash I'm like okay okay sign me in I I want to I want to be a part of it and it's project that's now in pilot stage and I'm going to be able to kind of guide it from zero to one along with of course many other wonderful people and professionals.

14:36 What's important at the stage when they're ready to make you an offer just to get more information from manager and like other people on the team, you want to make sure that it's going to be like you're going to be the right fit. And I ask a lot of questions like how does success look like in this role? What are some challenges? What are some problems? How do you work? What is important for you and what's not? This may seem to you like an overkill, but trust me, especially if you're just starting out, you don't want to be stuck in a job where you're unhappy for some reason.

15:09 So, it's like really important. And it's a good signal for a company as well. They look at you and like, "Oh my god, this guy is choosing as well. He he's not as simple. Like, if we just make an offer, he might say no." Because he has some criteria. And it's really important to show them that you wouldn't just go to the company that would make you an offer that you are selective. And I'm sincerely am selective because I want to be in a place where I'm happy. And after eight interviews and just couple more talks with managers and directors, I finally got an offer. It's the highest I ever received in my career. I'm going to be honest with you. it's higher than most of the offers that I seen on levels FII for the position. I'm going to be starting in the end of October and I'm just very happy that I got through. It was like a long road like 2 months, eight interviews references and everything. I'm going to share with you resources for preparation in the description for this video. And I wish you best of luck on your career journey.

16:19 It's hard. It's long. It's tedious. But I truly believe that you will manage. Thank you for watching. Subscribe if you like this video. And once again, best of luck. Bye-bye.