Ask yourself what's a business question you can answer. I have done some EDA- eg what times of day are the most orders made, do some drivers make more from tips than others Dashers (delivery people) have the freedom and flexibility to work when they want, while restaurants are empowered to reach a greater pool of customers. Restaurants have too many - or too few - drivers and can’t always manage capacity for burst demand. A data scientist will ask a few questions on how you crafted the solution and go through your thought process. If a signal is flagged as potentially risky, the team has the knowledge to better understand whether that’s typical of good users or if it’s something they should be concerned about. To scale with its order growth of 325 percent in 2019, DoorDash uses a 10TB Amazon Aurora Postgres cluster, Amazon ElastiCache, Amazon CloudWatch, Amazon Kinesis, and Amazon Redshift to provide real-time data analytics to its last-mile logistics network, merchant services, and customer membership program. Make the internet a safer place — Grow your career. Any help would be appreciated. I applied through a recruiter. This left DoorDash in a position of having to reimburse the victim (either directly or via chargeback) whose credit card was stolen after the victim disputed the charge. Hey there! In these early days of DoorDash, no automation was in place and most fraud prevention was done via manual review. Routing, scheduling, optimizing delivery queues for profit in various ways such as segmenting them, getting a better geographical understanding of tips, evaluating and ranking driver performance for performance reviews... You could even propose additional datasets to collect for specific purposes, such as turnover for correlation to delivery performance and tips, to evaluate whether pooling tips in teams or across the board would improve overall performance. In any position you are in. 5 Minute 'Big Data' Case Study: DoorDash https://www.linkedin.com/pulse/5-minute-big-data-case-study-doordash-adam-nathan. Practice data science interview questions from top tech companies delivered right to your inbox each weekday, 26 Oct 2020 – 6 min read, 31 Aug 2020 – Your information will be used to contact you about our service and subscribe you to our direct marketing communications. I'm delighted to engage with readers, clients and peers as we explore how to leverage advanced analytics to drive ROI and revolutionize our business models. Jobs. Interview. Dashers are faster and more nimble on electric bicycles. obviously your recommendations may be sorta out there since you have limited data. DoorDash: P2P Case Study Note: The following case study is an exercise in human centered design. DoorDash is the largest third-party delivery service in the world, supporting on-demand delivery for more than 340,000 local businesses and restaurants in 4,400 cities across the United States and Canada. Please see our Website Privacy Notice. The wealth of data that DoorDash has available to them in being part of Sift’s global network has been invaluable to the Risk team for not only identifying fraud but recognizing the behaviors of good users, as well. You know their business model, you know what data they have, you're expected to think of some ways to improve their business using the data, then execute. I have done some EDA- eg what times of day are the most orders made, do some drivers make more from tips than others. DoorDash interview details: 618 interview questions and 532 interview reviews posted anonymously by DoorDash interview candidates. The second segment will require you to write SQL queries and answer a few SQL questions. I am not affiliated with DoorDash in any way. "Some weeks I would just have one interview and other weeks around 7-8 interviews". DoorDash optimizes the timing of when a food deliverer - a “Dasher” - is sent to a given restaurant, taking into account a wealth of data points, including the restaurant’s track record, the historical prep time for each specific dish, current traffic patterns, the delivery vehicle type (e.g. The data science machine learning take-home challenge is also two parts. For as long as most data analysis and scientists without PhDs fail to deliver this minimum, we're going to keep being pressured to go for that PhD to further our career, just based on stereotype. DoorDash enables merchants to attract new customers and make more sales through our suite of marketing and full-service food delivery solutions. Our newsletter delivers industry trends, insights, and more. They're looking for some imagination... Come on, you don't need a PhD to think up some problem statements for a dataset like this. The user is prevented from returning to the platform, and the machine learning model recognizes that fraudulent behavior in other users. (3) The next step is the take-home challenge review call if you pass the assignment. they want to see how you can offer actionable insights to the business and use your data science knowledge to do so. Although these three teams are separate and work independently, in some cases they work very cross-collaboratively.