Microsoft Principal Data & Applied Scientist
Rutgers University Master of Science, Statistics
Current Time 0:00
/
Duration Time -:-
Progress: NaN%

How did you get to where you are today? What is your story? What incidents and experiences shaped your career path?

Summarized By: Arushi Chaudhary on Thu Feb 06 2020
My Ph.D. degree is in Industrial and Systems Engineering and my area is in Multi-variate statistics of modern or manufacturing processes. After that, when I was thinking about it I thought that I needed to do something interesting like some kind of medical stuff, so I went to Arizona State University to do my postdoctoral research. I went there and started working in neuro-engineering mostly, working on holding models of the neuron systems of non-human primates and animals. After that I stayed at Arizona University for four years and I decided that I would like to go into the industry because it is more exciting. I joined a small startup company which was a financial consulting company kind of a big stock financial consulting stuff more like a financial service company, they had different clients. The field models build financial solutions for them, like a customer's problem for detection or healthcare Problems like how do you know how many days a patient is going to stay in the hospital for years that's a critical issue for the Insurance company. I had a very interesting experience there actually I did not do any customer project but my manager put me to attend these kinds of operations such as scheduling. The reason I guess is that the company is benefited from these kinds of activities because we're small companies, they need to have some kind of check records in different problems. I spent about two years and I spent most of the attending those competitions and I did pretty well in some of the competitions. That's a really good experience for me, especially for my background, I work for the Statistics area but actually it is way more advanced than that. It's just right. So that really gave me an opportunity to polish my skills to get exposed to different moments such as recommendations such as feels care such as financials, etcetera. I spent about two years doing those compositions and think that I really need to have hands in the industry probably because the difficulty, the challenges that you are not going to be encountered in machine learning and machine learning compositions, in those compositions they have screening for you, most of them and the problem redefined to and all you want to do is to try different features you nearly missed already tried different models something that exactly improves your ranking but once in several scenes in the technical industry it really means that you need to be able to understand the business problem and try to translate the problem into some rational solution, that's something used in real life, they don't always prepared to you. You have to heal them and be able to handle the different types of data gives men have, which is messy. So I decided to draw on, Leave the company on the night on the Walmart labs, that's also very interesting when we're there I was not doing machine learning It gives me the opportunity to be exposed to be in the Walmart, every minute millions of users are coming shop and on the other hand, I was able to learn on the practice. I was leading a small team to pull some kind of internal Tool business intelligence called dashboards for the well being of use. After about 1/2 year, I'm still busy in my work, my life, machine learning and reading books. Learning was a kind of high tech company, so that time I got the opportunity to work for Microsoft and I moved to Microsoft in Europe in 14 and that's where I am now.

What are the responsibilities and decisions that you handle at work? Discuss weekly hours you spend in the office, for work travel, and working from home.

Based on experience at: Principal Data & Applied Scientist, Microsoft
Summarized By: Arushi Chaudhary on Thu Feb 06 2020
In my current team we are working with external capital partners of Microsoft, External partners are like I give you some examples. External partners we used to work with were big enterprises such as Nestle, Starbucks and Walmart, it can be any such company. We call them customers they are called customers of us or partners that you have no kind of problems with regular business operations. They feel because they have data, they have something that they want to build some kind of cloud-based solutions. They feel such solutions can be good for the business also this solution does not only include the machine learning model but also figure out how you can deploy this model into their production system so that they can use this on a regular basis. So this is my major responsibility, it's to work with partners, to go out to design a machine learning solution and tackle the business problem. We are supporting wherever the customer is from, so you can be from Europe, Asia, South Africa or South America, We are global worldwide. Many times we need to go fly into the customer's site, walk with them to discuss and understand their business problem very clearly and come up with a machine learning solution. So roughly we travel all over the world. I travel approximately 20 to 30% of the time. Microsoft office is pretty Flexible unless you have a meeting or you need to meet with someone personally, you have the freedom to work anywhere you want. You can work from the office, work from home, Starbucks or hotel. Some could start a tail or something like that, these days you know those kinds of virtual meetings such as Microsoft teams where you can meet people use all the using different virtual channels to meet with people within Microsoft or your external partners.

What tools (software programs, frameworks, models, algorithms, languages) do you use at work? Do you prefer certain tools more than the others? Why?

Based on experience at: Principal Data & Applied Scientist, Microsoft
Summarized By: Arushi Chaudhary on Thu Feb 06 2020
These days because of the machine learning in artificial intelligence is preparing a very popular and advanced solution for solving different problems. Before deep learning R and Python are the most common tools used by scientists and these are the most popular and talked about tools too. People don't send us their issues right off away, they look into the statistical backgrounds and particularly python is a much popular tool in people with an engineering background so python is more advanced than other tools. From that perspective we were mostly using python here and talking about deep learning people work here just for the experience. New tools and software are also coming up very quickly so it also depends on the tool most times it is also because of the way in which we work with partners, we also welcome young people collaborating with our partners. For example, we have won the project we have signed from me, but we're also expecting to take a census from the other side so you really would shoot the framework of the scenes that the partner is comfortable ways or his likes. Once the department becomes more familiar with us and you have the intention to learn python is an easy tool.

What things do you like about your job? Were there any pleasant surprises?

Based on experience at: Principal Data & Applied Scientist, Microsoft
Summarized By: Arushi Chaudhary on Thu Feb 06 2020
Our job is working with External partner supply, we were on a project for one partner for two, three or four months then we deliver their production supplies and we leave this project and then we jump into another part, so this means every two or three or four months, we always run into a new challenge. The challenge might be machine learning these might be similar to the business problems you always get these different challenges or different projects so that really makes me excited about the work. I am saying that our partner might be global, so you might get the chance to travel to meet people across the world that's another kind of exciting experience for you, your work is sponsoring you to travel across the world, That's a very good experience for your life.

What are the job titles of people you routinely work with inside and outside of your organization? What approaches do you find to be effective in working with them?

Based on experience at: Principal Data & Applied Scientist, Microsoft
Summarized By: Arushi Chaudhary on Thu Feb 06 2020
We work to resolve solutions for our external partners and here machine learning is just one component under the solution It includes the suggestions and questions in the machine learning model when it is making predictions it needs to get data from somewhere so you need to have a pipeline. They have a plan to get in customer assistance into the machine learning model so it can be consumed by the model that means many times we have to collaborate with developers and software engineers because they are ones who are responsible for building tools. When you are working you need to work with a project manager they're going to help you figure out the connection between yourself with Microsoft and the customers with Microsoft. It is done on a monthly basis there are three typical tables and the most efficient way to be working with them is that you work for the project we worked with, we have to go over to major groups of people who have actually coded with the developer. You need to make sure that both parties work efficiently and the best way to make that makes it happen is that both developers are dealing with scientists and show them your concerns. We got through with the model, so let them overcome this for once you have the interface it redefines on the post-party and work and that's mostly patient otherwise, the whole oppression is hopeless. 

What major challenges do you face in your job and how do you handle them? Can you discuss a few accomplishments?

Based on experience at: Principal Data & Applied Scientist, Microsoft
Summarized By: Arushi Chaudhary on Thu Feb 06 2020
The main challenge is to understand the business problem the customer is facing. We resolve and monitor the problems our partners and customers face once in every two to four months. Every time we have a new engagement or a new project we have to understand what are the customer's problem and the reason why he needs this kind of solution When you start and propose it wrong with the wrong machine learning solution the whole problem-solving process turns into a disaster and that is the first challenge that you need to overcome, you will encounter the second challenge as the process of problem-solving and the way or direction the problem needs to be solved sometimes you feel like you are right but based on my experience, there is no one way to tackle your problem you'll have to make some changes or you need to have good insight. If you have an insight the whole business problem can be understandable with the kind of screen algorithms you are using based on that understanding, for most of the problems you think of solving them with machine learning components but before actually solving the problem you have to know and understand the affected problematic area and you asked about my accomplishments so few of my accomplishments are that I have a published paper on acne problems that I have healed that happened while observing deconstruction to assess the diverting level based on self-images acne assessment. I had two patients that had to go through the largest operated vehicle that's some counseling and is not convenient for the customer. They're happy with one month or something even longer together, so now we have cellphones, smartphones the destruction will be all for the rest. Ms. Kane tells me that we feel the machine learning promotion you promised eversion so that the customer can just take the self-evaluation without indulging in the machine learning model. It is good to know that children have achieved so far.

What are the recent developments in the field? How significant are these improvements over past work? What are their implications for future research & industry applications, if any?

Based on experience at: Principal Data & Applied Scientist, Microsoft
Summarized By: Jeff Musk on Thu Feb 06 2020
When we talk about the recent developments and advancements in the field of machine learning and AI, Deep Learning has to be mentioned. It is growing very fast and there are a lot of new algorithms and applications coming out each day. It is changing and bringing AI and Machine Learning into a new different era. Traditional Machine Learning, you had to do handcrafted engineering to make the model work. Nowadays, because of configuration and Natural language processing, that burden is significantly reduced in those two areas. I feel that machine learning is being used in your workplaces too. These days you have to be very careful about how aware and competent your AI and your application are they should act and function responsibly you have to be able to protect the privacy of people who have shared their data with you for protection and also you have to be able to avoid discrimination model. There are very broad discussions regarding that kind of responsible AI because they have to be responsible and careful.

What was the hiring process like for your job? What were the roles of people who interviewed you? What kind of questions were asked?

Based on experience at: Principal Data & Applied Scientist, Microsoft
Summarized By: Arushi Chaudhary on Thu Feb 06 2020
The hiring process for this job was different for example we have openings and we'll go to HR people and the job post and the job description because of the different kinds of candidates we would have then the HR shortlist the candidates and their information like their country and all, there are several HR customs like do ou need sponsorship for your working position and what is your degree? where have you worked before and all so basically that are the kinds of requirements we require for the team. We have many Ph.D.'s, college graduates and masters people with us so everyone has this question about if we have some kind of summer internship opportunity with Microsoft will that hiring process be easier? of course, the summer internship interview and hiring process are roughly the same that's the turning point in the process we have a certain standard for a summer internship because If you can't get into the summer internship, we can't give you the opportunity to work but if you clear that we can work with you in the summer if you do well there you can pull one more summer internship or we can offer you a full-time job after you graduate this all is based on your performance. Now back to the full-time hiring process we keep a round of HR phone interviews that could scope down to smaller list people. The HR could have two-three more interviews just to check whether you have the right data sense and what is your background and all the selection is according to your capability. Most importantly, we look whether the person has data sense and is willing to work and that's more important than money because we will be like he's successful and the point is It's easy to collaborate with you. And also you have the right knowledge for this for this part mission.

What qualities does your team look for while hiring? How does your team interview candidates?

Based on experience at: Principal Data & Applied Scientist, Microsoft
Summarized By: Arushi Chaudhary on Thu Feb 06 2020
We check our candidate's data sense and writing sense by putting them into different scenarios. As I recall when I interviewed for my job after leaving Arizona university, the interviewer asked me questions to detect or identify my flaws she was like when you purchase goods from online you make sure that the online store is reliable so all this starts with your transaction. One strange thing happens when you purchase stuff online is there's no credit card they'll soon know what happened then you get a call from the credit card company you need to know why they are calling you we have to face and answer to many problems like that basically it is detecting the frauds so when you detect these frauds you have to put yourself in the shoes of the customers to have that experience and to know the feels. Even if you are in the production team you have to have a sense of fraud detection and what to do in a situation like that we need a conviction like that from our team.

What are different entry-level jobs and subsequent job pathways that can lead students to a position such as yours?

Based on experience at: Principal Data & Applied Scientist, Microsoft
Summarized By: Arushi Chaudhary on Thu Feb 06 2020
Data science has broad coverage and a vast horizon. The different jobs of data science are data engineer, data scientist, data analyst, data architect and many more. Your daily job could be like running some sequence or secret coding on the application like a data analyst and it a very important job at data science another job at data science is building machine learning models or supervising the models with your partners, well that's about it for the Microsoft. Some other incidents were like work It's more like a funding business which means with it sends provisions are facing more than just a portion of the tools that I didn't sense. Machinery models service in Microsoft Visibility Asia Freedom, those lectures stream all day there's another for the force, one would be something like being back home for Microsoft for research hundreds of searches so there are lots of other things and we are working there as the four different roles with scientists with Microsoft the difference between those four barrels that one's better to stream for us like the external basic right customers. In addition to those kinds of hard skills such as you should be able to understand the difference you have experience with expertise with industries, but such a refill just professional. We also need you to know who are we calling the calling skills. I also need you to be able to understand how to tackle that and help them. You feel the machinery solution goes out from other skills another and it will help you trace and hold to communicate with the customer and that is not a steal. I mean the authority won't move from the other three different types of roles, They start moving you off my current organization. You might have to take some time to build up your calling skills to please the customer that help you control the expectation of customers and that takes time.

What were the responsibilities and decisions that you handled at work? Discuss weekly hours you spent in the office, for work travel, and working from home.

Based on experience at: Staff Data Scientist, Walmart eCommerce
Summarized By: Jeff Musk on Thu Feb 06 2020
When I was working with Walmart I was leading a small team we were building the internal dashboards to help the merchants understand what items they need to prioritize. Merchants need to make sure and understand their customers they have to know what is in demand and what is trending for example easter is coming so they have to know what are the demands of people during easter at the Walmart account approximately five weeks earlier. There is a high competition in the market so we need to make sure that we are fully stocked with each item and that no item is out of stock we should see this as an opportunity. We have to keep a high database and records about all the queries people are making and searching their Walmart accounts to understand the trend, what items are popular and things like that. This helps us avoid wastage and smooth cash flow. The working hours at Walmart were very flexible, they had a work from the home policy so according to that you can do work from home every Tuesday or Thursday like you have two days to choose from and one reason behind this was the shortage of housing facilities there so if you live far from the town you can still work at home.

How did the program prepare you for your career? Think about faculty, resources, alumni, exposure & networking. What were the best parts?

Based on experience at: PhD, Industrial and Systems Engineering, Rutgers University
Summarized By: Arushi Chaudhary on Thu Feb 06 2020
I got my master's degree in the statistics department at Rutgers University. At that time we took ten courses before you can apply for a master's degree there so there were several courses there and they prepared me to have statistical thinking and some fundamentals for machine learning and I had a course called interpretation of model it means the relationship between the model and the dependent variable, it helps you to understand the interpretation result and how to interpret your results, how to interpret the coefficients of the models and what is the significance of the coefficients. It gives you statistical thinking of machine learning and statistical knowledge about the machine learning model.