Springboard recently launched their new Data Science Career Track, an online, mentor-driven course that promises graduates a job in the field or their tuition back! We chat with the Director of Data Science Education, Raj, to learn why he’s passionate about helping students make career changes, why their curriculum focuses on Python, and exactly how Springboard’s students are landing jobs when they graduate (hint: it’s not by blasting out resumes).
Tell us how you are involved with Springboard’s new Data Science Career Track.
I’m the Director of Data Science Education at Springboard, which means that my job is to create and maintain our data science curriculum, including launching new courses like our Data Science Career Track.
Why are you passionate about helping career changers become data scientists?
I actually changed careers through online education. I have a Master's and Ph.D. in Computer Science from Rice University. However, after a couple of years of working in the industry as a developer, I wanted to transition to data science. I tried out a data science class on Coursera, and then spent two years teaching myself through online classes at Stanford with their Continuing Education department, and then changed careers to data science. Before Springboard, I was Chief Data Scientist at a very prestigious startup in Atlanta called Pindrop Security. I had been mentoring for Springboard’s Python course, and when this opportunity came up to take over the data science curriculum, I jumped on it.
I’ve been through that career change using online education, so helping others and encouraging them to switch careers and upskill is something that I’m really passionate about. Working with Springboard is a way to have that kind of impact on a bigger scale.
What’s the difference between past Springboard courses and the new Data Science Career Track?
We teach three other Data Science courses: Foundations of Data Science (which is based in R), Data Science Advanced (which is based in Python), and Data Analytics for Business (which is based in Tableau and business case studies).
The Data Science Career Track is our first foray into providing career services. We've hired a full-time career services lead to take over the career aspects of the course, and I’m maintaining the technical curriculum.
What are the admissions requirements for the Data Science Career Track? Can someone start as a total beginner?
Because we offer a job guarantee for the Career Track, we’ve found that having some technical background does help students get jobs. So our admissions process involves a programming challenge and a statistics challenge.
For total beginners, we recommend one of our Foundation courses depending on their background. If they have no tech background whatsoever, then we typically recommend Data Analytics for Business. If they have a little bit of programming background or technical background, then we tend to recommend Foundations of Data science.
Will the coding challenge be in a specific language?
An applicant can choose their language, but the most common are Python, R, and Java.
Can you tell us a bit more about the Job Guarantee? What are the conditions of the job guarantee and why is it important to you at Springboard?
Roughly, our job guarantee means that students who complete all of our curriculum, including all the projects and follow the guidance of their career coach, and meet eligibility criteria are guaranteed a job within 6 months of completion. Eligibility criteria include willingness to live and work in one of 11 major US metropolitan areas, US work authorization, a Bachelor’s degree, and be age 18 or older. In addition, we require that to be eligible for the guarantee students are active in their job search and committed to their own professional success in that area. We have full confidence that if our students commit to the learning in the program, which includes both technical material and job search tasks, they will be successful in meeting their career goals.
Which data science languages have you incorporated into the curriculum?
We've decided to focus on Python. However, we don't really teach Python in this course; we assume that you know the basics of Python. We teach the Python data science stack, which starts with Pandas (a Python library to manipulate data, clean data, wrangle data), and then we teach Python libraries like NumPy, SciPy, scikit-learn for machine learning, Seaborn and Bokeh for visualization. In addition, we also cover Spark, which is one of the most in-demand tools for data engineering and scaling, along with PySpark (a Python interface to Spark) and MLlib, which is Spark’s machine learning toolkit.
Both R and Python are pretty common in data science, but if you're working as a data scientist, particularly if you're working on building machine learning algorithm prototypes, knowing Python is a huge advantage. You’ll find that R tends to be less connected to production systems, so we made a conscious decision to go with Python.
Did employer needs and feedback go into the curriculum design?
Yes. Many of our Data Science Intensive students were interviewing with employers, and we got feedback from those interviews. We had a sense of the gaps between our Data Science Intensive course and what employers were looking for. The Data Science Intensive was getting students 70% of where they need to be in terms of technical skills to find a job. So what was the remaining 30%? Employers said they wanted more experience with real world data sets and portfolios, and they also suggested we work on interviewing skills and job searching.
As a result, we weave career steps throughout our technical curriculum, so students are building their network, working on their LinkedIn profiles, and coming up with their pitch from Week One. Towards the end, when they're done with the technical curriculum, students can set up mock interviews. Some of our mentors have been interviewing candidates for many, many years, and they will give you feedback according to a preset rubric so that you get all the practice that you need for interviewing.
If you wait to finish the technical curriculum and then start your job search process, that's just going to cause a lot of delays.
What is the teaching style like at Springboard? What should students expect?
Our teaching model is completely online and self-paced, so students go at their own speed. First, you’re assigned a mentor, typically someone who currently works as a data scientist in the industry and has worked for a few years. They have not only data science experience, they also have the sense of what industry careers in data science are like.
Students work on the material in the curriculum at their own pace and the material is curated, which means that we collect the best content we can find on a specific topic. Then we assign mini-projects for each topic where students actually work on a realistic problem, and that's the way they learn each specific topic.
Throughout the course, they work on two capstone projects. One can be a little bit more foundational, the other might be more advanced. The capstone project should be as realistic as possible as you should use some kind of real-world data set, and the question that students choose to answer should have some real world value. Students need to write a proposal where they state the question, why they care about it, the value of the answer, and who the client is. In the real world, when you're working as a data scientist in industry, you're never working on a problem in isolation. You are typically working to prevent or solve a problem for a business client.
Knowing how to translate a business problem into a data problem and then communicating the results of your analysis back into a business context is a super important skill for data scientists. It’s highly underrated and something that employers always look for. The way we teach that skill at Springboard is by making sure that every capstone project they're working on has the client in mind. The analysis and deliverables should all be targeted for that client.
How do you keep students engaged while they're learning online?
That's a really good question, and this is something that many online education providers are trying to figure out. Assigning mentors is a big part of this for us, because it means that students are being held accountable. Students meet with their mentors once a week, online as we’ve built video calling into our platform. In their weekly calls, we encourage students and mentors to decide on goals for the following week.
Student advisors will also follow up to check in on students’ progress. If you haven’t made some progress over the last couple of weeks, the student advisor will reach out – that kind of human touch often helps many students. When students accomplish specific milestones, they have prompts to set up calls with their advisors. For example, once they update their LinkedIn profile, they have a call with a career adviser who will review their LinkedIn profile and give them feedback.
We’re also always thinking about how to better design our platform and curriculum to motivate students. For example, a lot of students are motivated by seeing their progress as they go through the curriculum, so we built those rewards into the platform. Student do well when they’re aware of their own learning style because they can work with their mentors and their student advisor to make sure we’re motivating them in the right way.
What have you found is the easiest way to land a job as a data scientist?
When you look for a job, especially in tech and data science, you often get the advice that you need to pump your resume full of keywords and then blast it out as widely as possible. That's really not the most effective way to find a job. Referrals are the way to find jobs in tech, and that means building out your network, and then using your network to find jobs, interviews, or referrals to companies that you've already done information gathering and research on.
Students need to be very strategic in the beginning before they send out a single resume. And we’ve built that idea into our career curriculum. For example, students may be required to find a major data science meetup near them, attend, and make five contacts, take five people out for coffee, or schedule an informational phone interview to learn about their company.
One of our students put his data science skills to the test and ran an experiment where he sent out hundreds of resumes to different job sites, and got an acknowledgement ~10% of the time. When he submitted applications through referrals, he got a phone interview 85-90% of the time.
The next class starts May 29th; how are the current students doing?
We’re teaching a couple of hundred students right now, and we have about 50 mentors in our network. Some of those students are getting close to graduation, and then will be focused on finding a job. We accept applications on a rolling basis: however, admissions are quite selective, with about only 18% of students enrolling after they’ve applied. Click here to see if you qualify!
Great, we can’t wait to talk to a graduate!