Since launching in 2013, Metis has always partnered with leading tech companies to design their bootcamps- they teamed up with world-renowned Rails shop Thoughtbot to develop their Web Development bootcamp, and partnered with Datascope Analytics to design their current Data Science bootcamp. While we’ve chatted with instructors and curriculum designers at Metis, this is Course Report’s first time talking with Metis students! In this Alumni Q&A, Emily and Itelina tell us what they were up to before Metis, their data science bootcamp experience and their lives post-graduation.
Liz: What were you up to before you started at Metis? Tell us about your education background, your last job, things like that.
Emily: Before Metis, I was working for a small branding firm where I was doing qualitative research and visual strategy/brand strategy, and I really liked doing research. I have a Bachelors of Fine Arts, which isn’t something you necessarily think of with data science. I got really interested in data visualization and wanted to learn more quantitative skills. I started taking courses in the evening in statistics, and that snowballed until I found out about Metis and this data science course and I thought, “Okay, I want to learn all of those things that they’re teaching.”
Itelina: I graduated with a degree in mathematical economics from Princeton University four years ago. In the last four years, I’ve been working as a management consultant at Price Waterhouse Coopers. I work mostly with healthcare clients such as hospitals and insurance and pharmaceutical companies.
I did a lot of projects that involved solving problems with data, so I was often the lead on the analytics work stream. I worked on analyzing data from hospitals and insurance companies and seeing different patterns in cost of care and quality of care. That got me really interested in solving problems in data, and I also did some of that in my undergrad degree.
I really wanted to find a program where I could the latest techniques, technology and tools, and Metis was the perfect program.
Liz: Itelina, what clicked and made you think, “I want to be a data scientist, go to a bootcamp and get that data science job after I graduate”?
Itelina: When I was an undergraduate doing economics research, I used statistical software to analyze data. When I was consulting, analyzing data wasn’t always the primary focus because I was part of a group working on strategy.
But we started working with clients who had increasingly larger data sets. Very early on I realized that some of the traditional tools used in data analysis, such as Excel, were obviously not powerful enough to do the kinds of analysis we needed with larger quantities of data.
So I started taking courses online forum Coursera. I took the data science specialization track and I learned a R and things like that. That made me realize that there are a lot of technologies and tools out there and that I wouldn’t be able to focus on learning all of those technologies while working at my job. That made me realize that I wanted to attend a bootcamp program.
Liz: Emily, you mentioned doing a night statistics class, right?
Emily: I did some online courses as well. Not through Coursera, but I did one that was through EdX.
Liz: Did you ever go to meetups or workshops before you thought about plunging into a full-time bootcamp?
Emily: I hadn't been to any meetups before. I did go to the open house that Metis had just to check them out to see if this was really something that I was interested in, but I wasn’t really going to meetups.
I did some in person courses as well as the course that I did through EdX. I was sold on the idea after I went to the Metis open house. After that, I researched the other boot camps in New York, and I felt like Metis really stood out to me.
Liz: Itelina, did you also look at data science bootcamps in New York?
Itelina: Yeah, I did. Honestly, I considered graduate school programs vs. bootcamps. I figured that doing the bootcamp would be good for me because it would be a shorter program, more practical. I looked at some of the other bootcamps in New York City. There’s a bootcamp called Data Incubator; they’re geared more towards students with PhDs.
There were a couple of other programs, but I just thought that Metis was great for what I was looking for because of its instructors, full day program and the curriculum, so that was why I chose it.
Liz: Emily, what were the other boot camps that you looked at?
Emily: I can’t remember the other ones that I looked at but I looked to see which ones were being offered in New York. The main reason I chose Metis is because they had their curriculum published. I think there was another bootcamp in New York that was teaching mainly R, and I was more interested in Python because I felt like it’s a language that’s very versatile and can be used for so many different things.
They touch on so many different languages like D3, which I wanted to get better at. We did a little bit of SQL, MongoDB and Hadoop. I wanted the full breadth of languages.
Liz: Tell us what the application was like? Did you have to do a coding challenge in a certain language?
Itelina: For the application process at Metis, you do a take-home challenge. You’re required to do it in Python but you can do it at your own pace at home. There’s a set of problems for you to solve with descriptive results and answers.
There was a video interview with one of the instructors. They do test your background but it wasn’t intense or difficult.
Itelina: I knew how to do data analysis in R but when I got the challenge from Metis, I said, “I need to learn how to use Python,” so I looked through a book and did the challenge.
Liz: How many people were in your cohort?
Itelina: I think about 20-25 people.
Emily: I think it was 21.
Liz: Did you find it to be a diverse class in terms of age, gender and race?
Itelina: In terms of gender, definitely. I think half of our class were girls and half were guys. Age, definitely; we had people who recently graduated from college and people who were trying to make a mid-career change. We had people who were born outside of the U.S., so definitely a lot of diversity in backgrounds.
Liz: I’m not super familiar with the world of data science but is it fairly male-dominated?
Emily: I’m not sure because I feel like it’s such an emerging field. That’s one of the things that I considered when I was looking for a job after graduation. There are many people trying to hire data scientists right now. I’ve met people who ask what I do and I say I’m a data scientist and they say, “My company’s been trying to hire a data scientist for months and months and they can’t find anyone.” That could be because there are more positions to fill, but it could also be that different companies are looking for someone very specific to fulfill their needs. I don’t know yet how if I feel that it’s a male dominated field; possibly.
Liz: Who were your instructors and was it important to you to research the instructors?
Emily: It was important to me. I was actually a bit disappointed because when I first went to the open house Irmak and Bo were instructors at the time. I met them and thought they both seemed really nice. I was excited for them to be my instructors then they weren’t. But, the instructors for our cohort were also pretty amazing.
Before it started, I looked at their LinkedIn profiles and Googled them to see what their backgrounds were. They were definitely very qualified.
Liz: Who were your instructors?
Emily: Erin Schumacher and Jon Hanke.
Liz: What was the teaching style like?
Itelina: The general format for the bootcamp is lectures for a couple of hours in the morning and some pair programming problems to improve our coding style. In the afternoon it’s mostly individual work on projects. The instructors are available and you can ask them any questions or ask you can ask your classmates. It’s half structured, half individual work.
Liz: You’ve both done a traditional 4-year degree. What did you think about the teaching style compared to a traditional university classroom?
Itelina: I thought the format was very innovative. It’s project based so every two to three weeks we do a project and make presentations. It’s very practical and simulates a real work environment where you’re delivering results in short time periods.
I thought it was a very innovative learning model because you’re learning and doing. At the same time, we have good coverage of the theory behind the different models and the technology tools.
Emily: I thought so too. Having the unstructured time in the afternoon — it was sometimes hard to stay focused, but it did replicate a real work environment in that you don’t always have someone looking over your shoulder and telling you “You need to do X, Y and Z.” You sometimes have to figure that out for yourself and manage your own time.
Liz: Were there also teaching assistants or former students that were around to help?
Itelina: We had one teaching assistant. She was around to help everybody with the problems sets as well as the questions so we had very good support.
Liz: Was there a good feedback loop with Metis?
Emily: We had one-on-one meetings with staff members including one of the co-founders. They like to do that a few weeks in to see how it’s going for you and to see what’s working for you and what isn’t, so that they can make real-time adjustments to the course. They encouraged us regularly. Occasionally they asked us to fill out anonymous surveys.
Itelina: I definitely agree. They were very active in getting feedback and making adjustments on the spot. If there was a particular topic that we wanted to cover more, they would schedule an ad hoc session in the afternoon during our free time. It was a very good process.
Liz: Did everybody that you started with finish with you?
Emily: I think we had one person who started who left after the first week. The other 21 people made it all the way through to the end.
Liz: Did you have tests or assessments throughout or was it mostly project based?
Itelina: We didn’t have tests; mostly project based assessments.
Liz: What technologies did you learn at Metis?
Emily: Python was the main language that we used. For different projects we were encouraged to use different technologies. SQL, Mongo DB, Hadoop, D3, Hive, Spark —we covered those and you could use them or not, depending on whether they were necessary for your project. Other people with more advanced knowledge coming into the course got into Neural Nets, but that wasn’t explicitly taught.
Liz: Emily, you mentioned your background and wanting to get into data visualization; were you able to do that at Metis?
Emily: You could go as far into D3 as you wanted to, but there was a lot of emphasis on communication. The three main things that they talk about are programming, statistics and communication. At the end of a project when you’ve done all the complicated statistical models, you need to explain it to your client or the stakeholders in the project. A lot of the communication comes from data visualization. That was discussed and we were encouraged to explore that.
Liz: Itelina, can you tell us about your favorite projects?
Itelina: For one of my projects I worked on some data that was published in a Kaggle competition. Kaggle is a company that hosts a lot of predictive modeling competitions. Sometimes a company will say, “We have this data set and we’re looking for someone to develop a predictive model on this outcome.” Kaggle will host this challenge as a competition and there is money involved, so people compete for these prizes. There’s a leaderboard where people see their scores. It’s a great resource for keeping up with the latest techniques that people are using as well as data sources.
I used Kaggle to get a set of data. That data set contained information on patients and their medical records. The topic was predicting a person’s risk for developing type 2 diabetes based on their past medical record. For that project I worked on developing a predictive model for a person's risk for diabetes based on these other factors. That was definitely an interesting project.
Liz: I love that because it’s clearly your former career influencing the way you use data science skills to solve problems that you couldn’t have solved before. Did you work on it alone or with a group?
Itelina: For that project I worked with a couple classmates who also picked healthcare related topics. Another classmate focused on the wellness aspect. She was analyzing diet and blood pressure, and how that correlated to their risk for certain diseases. Somebody else did something similar with heart disease. I worked with a group collaborating on similar topics but on different angles.
Liz: Emily, what was your favorite project?
Emily: It was probably my final project where I did some analysis of UN data. It was for a data visualization challenge that the UN was hosting on the Millennial Development Goals. It was a 15 year project in which the UN focused on specific goals.
One goal concerned curbing the spread of different diseases. I focused on HIV data, and found some interesting things going on with the spread of new cases of HIV in Sub-Saharan Africa. I looked at which countries were able to stop the spread of HIV.
It was hard to determine causation in the data, but I managed to come up with some interesting visualizations. I was a finalist in the competition. Also, the UN took notice of me and offered me a job!
Liz: What are you up to now?
Emily: I’m on a short term consulting contract with the UN right now. I’m working on a dashboard exploring some data related to climate change.
Liz: What was that process like?
Emily: I sent it to them because I had to make the entry into the data visualization challenge. Also, my school had a contact at the UN because they were hiring. I sent in an application and got an interview because they saw the work that I did for my final project and thought it was good.
Liz: I’m assuming you’re using data science skills for this project.
Emily: At first when they were telling me about the project I was a little bit scared. I knew it was only a three-month contract and I thought, “You want me to do all that analysis in three months? You want me to solve climate change in three months?” But they have a lot of analysts who’ve been working with this data for a couple of years. I am working with a team of people who are figuring out how to present it and build the dashboard.
Liz: What is it like transitioning from a bootcamp into a real data science position?
Emily: At first it was difficult because I’ve worked on so many of my projects individually. I had to figure out how to setup a workflow in terms of code.
On the other and, I really feel like I need to shake my imposter syndrome because I feel like I have so many skills. I’m working with people who are also very skilled, but I don’t feel like my skills are sub-par.
Liz: Itelina, what are you up to now since you’ve graduated?
Itelina: I’ve been doing a lot of job searching. I would say the process has been a lot better than I would have expected if I was just searching for jobs on my own because Metis provides a great network. We have great career services and great support.
In fact, before I graduated from Metis, I actually had a job offer. I think Metis really does a great job of making the right introductions and connecting you to the right companies that are a good fit for your skills.
I’m looking for a job where I can grow my career, so I want to take a little more time to talk to different companies and see what’s out there.
Liz: What’s your dream position?
Itelina: I definitely want to work for a company that has a really good vision, so a company that will make an impact on our society. I’m looking at opportunities in mid-stage startup companies, although not exclusively. I’m looking primarily for opportunities in healthcare, but I’m looking at opportunities in other industries as well. Definitely a data scientist/analyst type of position.
Liz: What is Metis’ approach to job preparation? Did you do mock interviews, resume building?
Emily: We did all of the above. They did a really great job. We had workshops on writing resumes and cover letters. We also did a mock interview where we were actually whiteboarding. We had the opportunity to do interviews for soft skills and one-on-one meetings with career staff.
Sometimes it felt like it was a lot to take in while learning so many things, but it was really supportive. I still communicate with the career staff. They still respond if I want them to look over a cover or some changes I’ve made to my resume.
Liz: Itelina, you’ve gone through a number of data science interviews with different companies. Have those been set up through Metis or through your own network?
Itelina: Partly through Metis and partly through my personal network. Generally Metis makes an introduction or tells us about a certain opportunity. We take it from there and see if there’s a fit.
Liz: What is a data science interview like?
Itelina: It’s actually very different from company to company. I’ve interviewed with companies where all that’s required is a take-home challenge then going to the company and meeting people and seeing if there’s the culture fit.
I’ve had other companies ask me to solve math problems. They would ask me to do that for an in person interview as well, and I would have to make a presentation to an entire group on a past project and defend my work.
I’ve had interviews where people asked me questions related to SQL. I think it’s different depending on what each position is asking for.
Liz: What has been the reception of you being a bootcamp graduate?
Itelina: People are generally interested. At almost every interview people ask me to describe what the program is like and the things that we do and how they teach. People are definitely interested in that kind of background. They’re interested in how well it prepares you for the kinds of problems you have to solve in a real job.
Liz: My last question that I ask everyone is was it worth the money and would you recommend it? Do you think that you could have learned everything you learned on your own?
Emily: I would say yes, absolutely. I think it was worth it. It was a challenge. It was really intense but I don’t think that I would’ve gotten all of this knowledge on my own, especially in this short amount of time.
I’m continuing to reinforce that knowledge and add to it with projects that I am working on. I really don’t feel like I could’ve downloaded all of that knowledge into myself in any other way.
Itelina: I absolutely agree with everything Emily said. I would definitely do it again in a heartbeat. It was definitely worth the money. The learning environment, having the type of classmates that you can learn from, the instructors — you definitely couldn’t have gotten it just by learning on your own.
Again the Metis hiring network is also really great, so I would definitely recommend it to anyone. I wish the program was longer so that I could stay there and learn more!
Liz: Anything you want to make sure people know about Metis, bootcamps or data science in general?
Emily: There is a lot of talk about data science because it is one of these fields where there are more open positions than there are people to fill them. I would say to do one of these courses, a masters or a bootcamp, you really have to love working with data, love a challenge and love problem-solving. You have to love working hard. I don’t think it’s something you can do just because you want a guaranteed position and a decent salary. You have to love this kind of work.
Itelina: I would say anybody who’s looking to get their foot in the door in data science or looking for a change should check out bootcamps. This type of program is very innovative. I benefitted greatly from the program so I think it’s great for somebody who is trying to start a new career.