Data Science Bootcamps

We recently caught up with Kim Nilsson, CEO at Pivigo, the company behind S2DS (Science to Data Science). We will be chatting about her transition from academia to Data Science education.We’ll also explore their process and curriculum at S2DS and some of the factors that differentiate them from the other European bootcamps.

Hi Kim, we appreciate you taking the time for this interview. Lets start with a couple of questions

Q : What’s your 1 minute bio / introduction?
A : I am an Astrophysics PhD, got my PhD eight years ago now and worked as a scientist for a number of years. When I wanted to leave academia and find a new career in industry, I struggled to know what to apply to and how to present myself and my skills. As a way out I came to the UK to do an MBA at Cranfield School of Management. It was there that the idea of Pivigo was born, together with my fellow MBA student and now co-founder Jason Muller. My experiences in the difficulty of the career transition, and his experiences starting a healthcare recruitment business led us to the idea of Pivigo. After the MBA I worked briefly in financial services before starting Pivigo with Jason.

Q : Can you tell us a little bit about your background and role at S2DS ?
A : Pivigo’s mission from the start was to support analytical PhDs in their transition into industry. As such, my role from the start was to talk to graduates and PhDs about their career options, and to see if we could help them. We realized early on that commercial work experience is everything, especially when applying for jobs in the UK, and we were inspired by other PhD-to-industry programs to create the S2DS program. As part of that we decided immediately that the program would have to include an element of real work experience, hence the company projects we do now.

My role today as a CEO is to overlook all the work we do at Pivigo, from reviewing applications and interviewing applicants, to setting up the events, mentoring teams and of course taking care of our business partners.  

Q : How did you get into Data Science Education / Training?
A : From a passion to support academics in their career transitions and realizing

• that data science careers are excellent and exciting career options for people like myself, and
• that even though they have fantastic skills they struggle to get hired due to the lack of commercial work experience.

The answer was clear; let’s create a way for these amazingly talented people to gain the experience they need, as quickly as possible, to allow them immediate career transitions.

Q : What is your definition of a Data Scientist?
A : For me, it is a combination of technical skills and personality. The technical skills we all know, it is maths, stats and programming. But the personality piece is at least as important. The Data Scientist needs to be a problem solver, needs to be curious and yet skeptical, needs to understand the scientific method and needs generally to be motivated by learning and developing. The final piece of the puzzle is communication skills; the ability to communicate complicated technical concepts to a non-technical audience is critical.

The by far most important aspect of the program is the project work which is completed in teams of three or four over five weeks, and the projects are real projects, with real data, and mentored by someone from our partner companies. As such, it gives a crucial work experience that makes all the difference when applying for jobs.

Thanks for the introduction. Now lets dig in a bit into how the program is structured and how you select prospective fellows 

Q : What’s the 1 minute bio / introduction on  S2DS ?
A : S2DS is a five-week, intensive immersion program for MScs and PhDs interested in a direct transition into a data science role. The program has three components: theoretical learning done via lectures and video tutorials, practical learning done in project work and networking aspects via events, panel debates and alumni meet-ups. The by far most important aspect of the program is the project work which is completed in teams of three or four over five weeks, and the projects are real projects, with real data, and mentored by someone from our partner companies. As such, it gives a crucial work experience that makes all the difference when applying for jobs. I would also highlight the diversity of S2DS, as we always have a good mix of nationalities, gender and disciplinary backgrounds.

The final thing to note is that we run two events; the London event and our Virtual program. The London event runs in London in August annually and takes circa 80 – 100 participants, whereas the Virtual event runs several times per year and takes circa 15 per program.

Q : How did the idea for S2DS come about and what do you hope to achieve with it?
A : Jason and I were eager to support scientists like myself to make the transition, but we acknowledged that the typical lack of work experience was holding them back from being hired. I read about the data science bootcamps that were being set up in the US and thought that was a perfect way to create an environment where companies looking to hire, and PhDs interested in new careers, could meet. After some careful thought around how we wanted this program to look, we launched S2DS.

Whereas we initially had only the interests of the scientists in mind, we now see that we serve multiple purposes. We have clear evidence that S2DS does get our participants into new jobs and new careers, but we have also seen that the partner companies gain enormous benefits from partaking. Our vision for the future is now to serve both sides; to help the career transitions of analytical academics, and to support companies who wish to gain insight from their data.

Q : How do you screen and select students / fellows for your program?

A : There is an initial written application, which includes uploading a two-page CV and, for the London program specifically, to write a few short essays. Following initial screening of the written application, a fraction is invited for phone or Skype interviews with one of the Pivigo Team. The interviews are not very strict, we like to get to know the person and their motivations. We then select applicants based on the combination of the written application and the interview.

Q : Can you describe the typical background (academic / professional) you look for in your students / fellows ?
A : I’m not sure there is one! Minimum requirements are a PhD (for London event) or an MSc (for the Virtual) in an analytical topic, strong maths/stats skills, intermediate programming skills in a mainstream language and great communication skills. Other than that I think we have taken people with some 20 different disciplinary backgrounds, with no work experience, only academic work experience or some commercial experience. Ages have ranged from 25 to 45. The cohorts are really very diverse but what they have in common are great skills and super strong motivation and drive to start a career in data science.

Q : What type of skills or traits  do you look for in a prospective student / fellow?
A : Beyond the technical skills mentioned earlier (maths/stats/programming) we look for strong communicators who can explain technical concepts really easily. We also look for people who are really motivated to make the transition, and who live and love to learn. A great sign, for example, are when the applications have already taken courses, joined hackathons or solved challenges before joining S2DS, as it really shows motivation. Finally, we like people who are friendly and collaborative. Data science is a team effort, not a competition, and we want team players.

The fact that they have a real project to talk about, showing they understand what it is like working in a commercial environment, is absolutely key to securing the job. Beyond that we have special career coaching sessions, and give the Fellows any and all support we can give after the program finishes, including introductions to hiring companies.

Thanks for that run down. Now lets move to placement and how your fellows transition into industry careers

Q : How many cohorts have you gone through?
A : We have completed two London-based programs and one Virtual program by now.

Q : Do you run multiple cohorts at the same time?
A : Not yet, but we may do so in the future.

Q : What is your typical cohort size?
A : It depends on the program. The London-based program takes somewhere around 90 in each year, the virtual programs are smaller, around 15 per cohort.

Q : Can you share what your placement numbers look like for your most recent cohort?
A : For the London program that finished in September 2015, 60% were in data jobs three months after the program ended. If you discount those that are still finishing their PhD degrees, i.e. look at only those who are actively job hunting, it is closer to 80%.

Q : Can you also share your historical placement rate (within three months of graduation of each cohort)?
A : It is similar for both years.

Q : What percentage of your fellows eventually get Data Scientist vs Data Analyst vs Other  technical jobs?
A : That’s a tricky question. They do end up with all kinds of job titles. Some also return to academia, realizing this path was not right for them or they are not ready yet to leave academia. It is definitely a majority that go into Data Science roles, a small minority go back to academia and the remainder split between Data Analyst roles and other engineering roles.

Q : Can you share with us where some of your graduates work or will work?
A : They have gone into a lot of companies! From very large ones like Royal Mail, KPMG, Facebook, Google, British Gas, Marks & Spencer, King, Deloitte etc., to small start-ups and everything in between.

Q : Can you share with us what industries your graduates work in?
A : The most prevalent sectors are Technology, Retail/E-commerce and Consultancies.

Q : How do you prepare your fellows to be very competitive for Data Science jobs?
A : The main preparation is the work experience they get in their project work. We hear over and over again how Fellows have secured their jobs by talking about their S2DS project experiences in interviews. The fact that they have a real project to talk about, showing they understand what it is like working in a commercial environment, is absolutely key to securing the job. Beyond that we have special career coaching sessions, and give the Fellows any and all support we can give after the program finishes, including introductions to hiring companies.

Q : Do you have a hiring day and what percent of students are typically placed from a company they meet at hiring day?

A : We do hold a job fair at the end of the program, where companies come in and present themselves. Some of the introductions following this summer’s program are still pending, but some 30% of the ones who got jobs since September got them via the job fair.

Q : For organizations looking to hire Data Scientists what should they look for..Ivy Degrees, PhDs, Extensive experience, Quantitative Background, grit and determination?
A : Having the right technical skills are obvious, but more than that I think it is important to find those with the right personality and drive. It is absolutely not about exactly what degree you have, or where you got it. It is also not necessarily about what experience you have. What I find most important of all is to find the people with the right mindset. If your next hires have motivation, drive, curiosity and maturity, then they will learn whatever skill or tool they need to learn to complete a task successfully!

Of course the hottest topic right now is deep learning, and the hottest tool Spark, but that doesn’t mean that I think any aspiring data scientist should rush out and learn about these. In fact, I find that it is much more important to master the basics; say, pick a language like Python and learn it well, learn how to solve simple decision trees, regressions or naive Bayes problems.

Thanks for shedding some light on that. Now lets talk about the culture, learning environment  and curriculum at S2DS

Q : Can you give a short summary of a typical day in the life / week in the life for your fellows?
A : Oh, tricky question. A typical day probably starts with meeting up with the team in the morning to talk about the latest progress, and what is going to happen that day in the project work. Then off to working on the data science project, with occasional coffee breaks and interspersed with conversations with other cohort members about their work. In the afternoon there may be an optional presentation by, for example, one of our alumni. Or there could be a meeting with either the company mentor or their internal Pivigo mentor to discuss progress. In the evening we may have one of our company events, when a partner company invites the cohort for some presentations and drinks, nibbles and networking.

Q : How do you improve your process and instruction from cohort to cohort at  S2DS ?
A : We ask every team to give us direct feed-back during the program, and we also send out an anonymous feed-back form to all the participants after the program, to canvass any suggestions for improvements we can get. The feed-back is very important to us and we made a lot of changes, for example, between the first and the second program based on the feed-back from the first year. We also felt those changes really improved the program, so it is extremely valuable to us!

Q : What skills and tools do you think should be emphasized more in Data Science education?
A : There is so much emphasis on machine-learning and algorithms today, but no one tells you before you get into it that 70-80% of your time as a data scientist will be spent on cleaning and preparing data for analysis. This is a very “un-sexy” topic, and so it is often neglected. Other than that, real world experience is critical, as well as communication skills and also some leadership skills.

Q : Given the very fast progression in the field, what skills do you think will be most important for Data Scientists in the next few years?
A : Of course the hottest topic right now is deep learning, and the hottest tool Spark, but that doesn’t mean that I think any aspiring data scientist should rush out and learn about these. In fact, I find that it is much more important to master the basics; say, pick a language like Python and learn it well, learn how to solve simple decision trees, regressions or naive Bayes problems. If you have mastered the basics, learning new languages and methods is easy. Having strong basic skills and a will and aptitude to learn fast will be very important going forward.

Q : The Data Science bootcamp space is getting quite crowded, how does  S2DS differentiate itself?
A : Size and quality of network, diversity and real company project work I would say. Very few bootcamps offer real company project work, and if you add in the diversity of our program (both on the participants, and the company side!) and the strength of the network you are included into post-S2DS, I think we are unique. There are already 180 Fellows in our alumni group, most of which are in data science jobs in Europe or globally. As a Fellow from our program you become a lifelong member of this group which is a significant perk.

Q : How do you work with your students to ensure they’re assimilating most of the material in a very short period of time?
A : There are two critical aspects to S2DS that support this. First, learning by doing. Almost all of the learning is in the projects, which is real, hands-on work. If you solve a problem once, you know how to solve it and similar ones forever. Second, we try to create an innovative bubble at S2DS. Whether on campus or on the Virtual program, we frame the days within the program and cohort to avoid external distractions. It seems to work!

Q : How do you help students deal with burn-out?
A : S2DS is a relatively short program, although it is of course very intense. We encourage social get togethers, as a way to blow off steam. We also never schedule any official work on the week-ends. There have been one or two cases of participants taken ill, or struggling with energy levels during the program, and we support them as much as we can. What we have actually seen is that they are extremely well supported also by their fellow team and cohort members, who all look after each other and pick up the slack.

Q : Are students ever asked to leave or are kicked out mid-way through the course or at anytime during the course?
A : It has not happened yet. We have been fortunate to have selected excellent participants for the program, and they have all contributed a great deal to their projects and their wider cohorts. If there are any personal issues within a team, we try to catch it early in our mentor meetings and work it out. It is in principle a realistic experience, to be forced to work with people you do not know.

Q : Can you give us a sample of the tools, languages and techniques  your fellows are exposed to during the course ?
A : It varies a lot depending on what project they work. We give an overview on standard tools and languages such as Python, R, Java, Hadoop etc. but the main learning in the projects will be unique to each team. Each project uses the tools and technologies that the partner companies sets as preferences, although in some cases the companies have no preferences and the teams end up choosing their own set-up. Overall, most teams work with Python and SQL, with a not insignificant fraction of teams also using R and Excel.

Q : On a typical day, how much time do your fellows usually spend in formal lectures, working on problem sets, listening to guest speakers / networking , etc   ?
A : The program varies a lot during the five weeks, but averaged out across

the program I would expect circa 75% hours spent on project work, 15% on lectures and guest speakers and 10% on networking.

If you come from an academic background and wish to become a data scientist; have confidence! The one thing I see over and over again with academics that absolutely kills me is the lack of confidence. Applicants and participants who just do not believe their skills or experience are valuable.

It is nice to see S2DS has identified lack a experience  as one of the impediments for academics making the transition to Data Science and then work with fellows to address that. Lets wrap up with a few more questions

Q : What do you feel is broken with Data Science education in general and how is S2DS  trying to fix it ?
A : In the UK, data science education has been non-existent up until very recently when a few Universities have started dedicated data science careers. So first step, create this education! Secondly, I think that data science in a commercial context can never fully be learned in a theoretical environment. It has to involve real, hands-on experience. You would never expect to let loose an engineer on a machine with only text book knowledge. The same applies to data science. It has to be a practical degree. Finally, a problem that Universities will always struggle with is the rapid change in best practice in data science. The tools used in industry can change as rapidly as every six months, and Universities are not capable of changing their curriculum at such a short pace. S2DS solves these problems by creating the internship-style project opportunity, and by tweaking the curriculum between each cohort.

Q : What problems in Data Science keep you up at night?
A : How to get more companies to see the light, and want to hire data scientists! There is still a lot of hesitation in UK companies, whether there is enough value in their data to warrant an investment in people. I try to tell them that you will have to get started sooner or later, why not take the competitive advantage of being an early adopter?

Q : Have you faced any major challenges in running  S2DS  ?
A : The greatest challenge for us was to get the sponsorship we needed to cover the costs of the first program. Because we wanted to attract participants from all over the World, we knew we had to provide a campus with e.g. free accommodation. But we also did not want to charge the participants a huge amount of money as we knew that many would not afford it. Hence, we had to find a number of company sponsors to cover the costs. The challenge was that because no one had done it before, companies were reluctant to sponsor. We were grateful to have a prestigious company like KPMG believe in the concept and become our Principal Partner for 2014, thereby securing the support we needed to run the first program which, of course, ended up being a great success.

Q : What markets / verticals are you currently focused on ?
A : Most of our partner companies are UK-based, and our focus geographically is limited to Europe at the moment. In terms of sectors, we have no particular focus. We are actually very proud that our partner companies come from a diverse set of sectors (Technology, Retail/E-commerce, Healthcare, Consultancy, Fintech, Not-for-profits etc.), as it means there is a project for everyone, and there is more shared learning when projects from different sectors get solved side by side.

Q : How do you feel the European job market differs from the US market?
A : I am not that familiar with the US market, so it is hard for me to judge. Even within Europe there are large differences in hiring practices and market maturity. In the UK especially, experience is everything. That is why ex-academics find it so hard to get their first job, as they typically have no commercial work experience. In Germany, where I worked myself for six years, titles and education are more important. Southern European countries still struggle with their recovery from the financial crisis, and hiring is slow. Scandinavian/Baltic countries are much at the forefront of tech development, but is a small market.

Q : Who are the Data Scientists that inspire you?
A : I have a number of people I call my ‘data heroes’, and they are typically not famous or well-known. They are typically mid-level managers in large organisations, fighting in the corner of data science. They know that there is huge value to be gained, and they work so hard to prove this to the upper levels of management. They usually succeed, and become very successful, but it takes dedication and a lot of very hard work.

Q : Any parting words for prospective Data Science students or Data Scientists that are just starting their careers?
A : If you come from an academic background and wish to become a data scientist; have confidence! The one thing I see over and over again with academics that absolutely kills me is the lack of confidence. Applicants and participants who just do not believe their skills or experience are valuable. And yet, they get incredible compliments from the partner companies once they are on the projects. I do believe this is one of the main reasons why academics struggle to get their first job offer, because they are not confident enough in their abilities.

If you are just starting a new data science career; have patience! Data science is still a new discipline, a change within each organisation. And change within an organisation is painful and takes careful management. Some employees will struggle to keep up, and others will resist. Be patient, carefully explain your work to anyone who should care and never show your frustrations. And enjoy the ride.

Thanks again Kim for sharing some of your thoughts with us. We do appreciate it .

To find out more about S2DS you can either reach out to Kim Nilsson, engage with S2DS on twitter @S2DS_School or reach out to their former students or Instructors. S2DS is currently enrolling for their Virtual session in March 2016 and their London session starting August 2016

Also, please stay tuned for the other Data Science Bootcamps Founder Interviews we have in the pipeline at Data Science Bootcamp Founders Interview Series