Bringing Data Science Ambitions to Life

The Data Incubator is a leading education resource that equips professionals with the skills they need to succeed in the data science industry. With over 600 companies hiring their graduates, an 88% employment rate within six months, and at least 128k dollars in compensation, TDI has established itself as a valuable bootcamp and placement company! 

To help potential students better understand the program, we held an panel discussion with TDI alumni Jonathan Burley, Marcos Huerta, Matt Chenn, and Robert Schabinger to discuss their experiences after completing the program. 

In this discussion, the graduates shared their thoughts on how TDI prepared them for the industry and helped them advance their careers.

Background of panelists

Jonathan Burley

Jonathan Burley completed his Ph.D. in Computational Geophysics and Climate at Oxford University, where he was awarded the Guralp Prize. His Ph.D. research in coding and computational climate laid the foundation for his career in data science. Jonathan honed his skills at the Foundry studio, working with Fortune 500 companies to prototype AI SaaS solutions. Currently, he serves as the Head of Data Science and Chief Data Officer at Actifai, a startup that optimizes offers for cable, media, and telecom providers. As a founding member, Jonathan’s unique background and expertise make him a driving force in the data science community.

Marcos Huerta

Marcos Huerta holds a Ph.D. in Astrophysics from Rice University and spent almost a decade in science policy before transitioning to data science through the TDI program in the fall of 2018. Today, Marcos serves as a Manager of Data Science at CarMax in Richmond, VA, where he started as a Senior Data Scientist focusing on pricing algorithms and systems for appraisal products. Marcos’ background in both science policy and astrophysics has provided him with a unique perspective on the challenges faced by data scientists, positioning him to make a significant impact in the field.

Matt Chen

Matt Chen’s academic journey began with a background in environmental engineering and chemistry, followed by experience as an academic administrator and research faculty. After completing the TDI program in the Summer 2020 Cohort, Matt transitioned to data science and now serves as the lead data scientist for the Xfinity Flex product team at Comcast. Matt’s personal projects include a web app that uses a convolutional neural network to generate string quartet arrangements of sheet music. His diverse experiences in academia and data science allow Matt to approach complex problems with a unique and innovative perspective.

Robert Schabinger

Robert Schabinger, a former high-energy particle physics researcher, entered the data science field by participating in The Data Incubator’s Winter 2021 Cohort. There, he developed a web application for summarizing Amazon product reviews in real time. Robert now works as a Data Scientist at DCI Solutions, focusing on using machine learning to develop cybersecurity solutions for military contractors. Though the details of his current projects are confidential, Robert’s background in theoretical physics and data science enables him to contribute meaningfully to the Department of Defense’s initiatives, making a significant impact in the field.

Here are their responses to some critical questions about their experience with TDI. 

Can you tell us about your current role and briefly describe what that looks like at your company?  

Marcos: As part of CarMax’s core pricing systems, I specialize in the online Appraisal lane. Previously, CarMax used algorithms to determine a fair offer when a customer brought their car to the store, but we now conduct many appraisals online. 

Since joining the team, I have primarily focused on developing the systems and algorithms behind this product. If you visit carmax.com/sellyourcar, you can use our online appraisal tool to receive an offer in just a few seconds, powered by our computer algorithms. My team and I are responsible for developing the algorithms and systems, as well as managing the cloud infrastructure and code that supports them.

 Jonathan: I work at Actifai, an AI SaaS company that helps telecoms make the best first offer to their customers. It’s a bit of a tricky sale since the agents don’t know much about the customers who call in. That’s where our AI system comes in – it quickly puts together a model based on the available information to make the best initial offer and suggest additional products. 

My day-to-day work involves building the product, monitoring production systems, and constantly innovating the model to improve it using reinforcement learning. I’ve come a long way from being the company’s first data scientist to now managing a team of six. These days, I spend more time thinking about data strategy and allocating resources to tackle the company’s challenges.

Matt : I work for Comcast, and I’m part of the entertainment products team that deals with cable products, streaming boxes, and over-the-top streaming services. My role involves working closely with the product development teams to make data-informed decisions about new features or improvements to existing ones.

There are three main aspects to my job. 

Firstly, I spend a lot of time providing general product analytics. This includes tracking how people use the product and what they do with it. Secondly, we do traditional data science tasks such as creating predictive inference models that help identify what drives product performance or customer engagement. Finally, we help design and implement A/B testing, which helps establish causal effects and improve customer engagement.

Robert: I’m currently working as a senior data scientist, but my role is quite diverse. Although data science makes up around 20% of my work, the rest involves a mix of software development, mathematics, and algorithm development. I enjoy finding innovative ways to solve complex problems by exploring research literature and utilizing reinforcement learning techniques.

What made you decide to join The Data Incubator?

Marcos: When I found myself between jobs, it was an easy decision to join The Data Incubator program. I had previously worked in science policy in DC for a long time, but I wanted to transition to data science and find something that allowed me to work remotely from anywhere in the world. I had taught myself some Python but wanted to refresh my technical skills and learn new tools. TDI was a perfect fit.

Robert: I was ready to make a career transition. It happened to be around the time when the Insights program and The Data Incubator were both available. Insight, unfortunately, had some issues and became less visible in the industry, making the decision to choose the Data Incubator a no-brainer. At the time, both Insight and the Data Incubator were considered the most prestigious programs for career transitions, especially in fields like physics. So, for me, it was a natural decision to pursue data science. And every time I reflect on my experience, I can confidently say that choosing TDIwas the best investment I have ever made.

Jonathan: These days, a lot of physicists are transitioning into data science roles. In fact, I’d say about three-quarters of my undergrad physics peers are now working in data science across various companies. Personally, I knew I wanted to go into data science after finishing my Ph.D. and moving to DC with my partner. When considering my options, TDI stood out as a prestigious and selective program that would be a great first step into a full-time data scientist role.

Marcos: During my graduate studies, I spent most of my time in a chemistry lab doing wet lab work. We did have some quantitative coursework, but it was far from what is expected of a traditional data science candidate. So, I felt that I needed more structured instruction to transition into data science. 

While there are many online resources and books available, I found that programs like TDI offered valuable face-to-face interaction with experts and tailored instruction, which was very helpful for me. I applied to TDI because I believed it would be useful, and it turned out to be the right choice. 

Can you share a bit more about your experience with The Data Incubator and how the program has impacted your career development?

Marcos: The best thing about The Data Incubator was the series of Capstone projects. Each week, we were given a new project, like web scraping or machine learning, and we had to learn it in a week. It was challenging, but by the end of the week, I felt comfortable with the basics. This experience taught me I could learn quickly and gave me the confidence to tackle new challenges. In data science, things move fast, and being able to learn quickly is crucial. TDI helped me develop that skill, which was the most valuable thing I gained from the program.

Jonathan: Marcos’ response to the previous question is spot on. Data science is a vast field, and most jobs only cover a subset of it. However, TDI’s program does an excellent job of covering a broad range of topics in a structured manner, making it easier for learners to remember and apply the concepts. The program provides plenty of hands-on experience, making it an ideal choice for anyone who is willing to learn and put in the effort. Moreover, the program offers opportunities to discuss specific topics with instructors, making it a faster and more efficient way to upskill than trying to learn independently. Therefore, I highly recommend the Data Incubator program for its breadth of material and specialized instructors who provide extension opportunities.

Robert: I want to emphasize the fantastic network of hiring partners that The Data Incubator has. It was really helpful for me to overcome the imposter syndrome. I realized that many people appreciated my skills, despite not learning everything about data science in an eight-week course. The program helped me develop a keen hunger to excel, learn, and contribute to businesses. I think this aspect of the program should be highlighted because it’s different from what other people experience. 

I had a friend who applied to 50 places, got rejected from all of them, and only had one interview. But after completing TDI, he landed a job with Revellio Labs. I recommended TDI because I knew it would work out for him. The program’s hiring partners are invaluable in helping you secure a job in the data science field.

How challenging was the program? Did you ever feel overwhelmed? 

Matt: It really depends on your fundamental understanding of data science concepts and how familiar you are with Python. Python is the language you’ll be using for projects and capstones. Having some knowledge and familiarity with the language and concepts will help you avoid playing catch-up each week. The challenge level is appropriate; it’s not too easy or difficult. There were times when I had to seek help from peers, instructors, or online resources. It was never impossible, but it did require hard work.

Robert: I completely agree. The program was appropriately challenging for me too. In hindsight, I would have given myself more time to prepare for the course, maybe three weeks instead of twelve days, since I didn’t have much knowledge of Python at that time. However, it wasn’t a significant issue, and the course materials, including the Python scripts and IPI notebooks, were very helpful and offered a lot of guidance. Overall, the experience was challenging, but in a good way.

Marcos: When I did the program, the first week was intense and overwhelming, but it got better as I found my rhythm. Balancing time management, weekly projects, capstone projects, interviews, and resume is doable. I took the full-time course, which meant some full days and sometimes a flooded email inbox. Learning Python and data science simultaneously can be harder, so if you don’t know Python, there’s a 12-day crash course that can help. I knew some Python from my background in data science with R, but I struggled between the two worlds.

Is there anything that you use in your current job that you learned at TDI? 

Robert: Python is a big one for me, I would say. While I don’t use ScikitLearn every day, I think it’s important to learn about these basic toolkits for one’s development. It was critical to learn the core concepts, especially since I didn’t really know what machine learning was before. Having the fundamentals was extremely helpful. Although I don’t use Beautiful Soup on a daily basis, it’s great to know how to use it because it may come in handy one day. 

Matt: The program was short but intense, and it was difficult for me to fully grasp everything that was covered. The instructors were skilled at selecting key topics to focus on, but inevitably, some materials were left uncovered due to time constraints. 

While I was able to retain a lot of what was taught, some of it may be outdated now, as new versions are constantly being released. For instance, I use Spark frequently, but the module we covered in the program is outdated because the API was updated soon after. This is why I believe that understanding the concepts and immersing oneself in the tool ecosystem is more valuable than solely relying on specific tools. Tools constantly evolve, with new functionalities and changes in API. However, the program helped me understand what’s available and how to keep up with the new tools and their intended uses.

Robert: Absolutely, Matt. Saving all the IPI notebooks is a great idea because they can come in handy later on. It has definitely happened to me where I needed to reference a past snippet, so having them all in one place is super useful. And don’t forget to save all the problem sets too, as they can be great resources to review and practice with. 

Jonathan: The program’s breadth covers important fundamentals that are always useful in day-to-day work. I find myself using the program most often when interviewing new data science candidates. I rely on doing deep dives on particular topics or skimming across several subjects to test if someone can keep up with me. I may run through two weeks of content from TDI during an interview to test how well someone has learned data science before they try to convince me they can be a data scientist. 

Looking back, do you have any advice that you would have told yourself when you were going through the program?

Robert: Take coding challenges seriously! It can be extremely useful in preparing for job interviews, not just in data science but also in software development. While learning the syntax of different languages is important, understanding dynamic programming and the way to think about problem-solving can greatly benefit you in the job market. Although it may be challenging to do them in real time, saving and revisiting coding challenges can be helpful in the future. While it’s possible to get a job without a mastery of Python, continued learning and improvement are important in the tech industry. It can be difficult to find the time for self-improvement, but investing in your skills is a worthwhile endeavor.

Marcos: In my experience at TDI, I found that paying attention to the hiring partner’s interview style was crucial. I had a book from TDI that provided intel on what to expect during the interviews. Some partners might expect coding challenges, while others might not. Take every interview seriously because you never know which company you might end up with. Even if there are companies you don’t know much about, prepare for each interview as best you can and know what to expect. Be aware of what’s coming and be prepared, especially with Python skills, as some companies might require it.

Matt: When it comes to the capstone project, I have two pieces of advice. Firstly, start thinking about the topic or project as soon as you’re accepted into the program. Don’t procrastinate like I did. Make a list of ideas, even if they seem crazy. This will help you find a project that showcases not only your abilities but also something you’re passionate about.

The second piece of advice is to prepare for your project demo. Even if you usually like to wing it, don’t do that this time. Write out a script word for word so you can make the most of your tight timeline. Practice speaking with cadence and enthusiasm, so you’re not monotone. This was specific advice to myself, but many in my cohort found it helpful. So, take these tips to heart and make the most out of your capstone project.

Jonathan: As someone who has gone through the program, I believe that the key challenge is the large volume of content and questioning one’s background. To prepare for the course, it’s important to identify what might hold you back. Do you need more business knowledge, Python-specific knowledge, or a better understanding of statistics? These are the three most important areas to focus on because data science is just a new slang term for doing statistics at scale and quickly.

Another important aspect is your GitHub page. It’s surprisingly important when applying for jobs. If you’re putting your code on GitHub, it should be presented professionally, meet coding standards, and include appropriate commenting and data types. It’s worth tidying up your GitHub after attending The Data Incubator program  and prioritizing projects demonstrating a genuine interest and unique approach. Remember, first impressions are important, and having a well-organized and professional-looking GitHub page can make a difference in getting hired. 

Marcos: I completely agree with the importance of having a good GitHub page. While it’s not the only factor, it can definitely make a difference if you’re linking to it on your resume. You don’t want people to click on it and find a repository you’ve never touched or something too basic. Adding a polished Data Science notebook or another unique project can make your page more impressive. If you don’t have anything to showcase, it’s better to not include it on your resume at all. So, ensure there’s something worthwhile to see when someone clicks on your GitHub link.

What do you look for when hiring data scientists?

Marcos: As someone who reviews resumes and conducts interviews, I agree with Jonathan’s earlier point that having business sense is helpful in applying statistical knowledge to real-world problems. However, what we’re really looking for in candidates is analytical ability and problem-solving skills.

It’s great if you have experience with Python and a well-crafted GitHub page, but what really stands out is evidence that you’ve applied your analytical and quantitative abilities to data science problems. This could be demonstrated through an interesting project you’ve worked on and what it accomplished. If you have relevant work experience or a PhD, that can also be helpful. Ultimately, the resume gets you to the interview, and then it’s up to you to perform well. TDI participants will likely have plenty of interesting projects to showcase their skills. 

Jonathan: Just to provide some context, I’ve been involved in hiring data scientists, having worked at Foundry. I’ve interviewed a couple of hundred candidates during my time there, and I can tell you that there is no one-size-fits-all definition of a data scientist. It really depends on what the company is looking for and what business problem they want to solve.

When it comes to hiring data scientists, the ability to think strategically and logically is key. You need to be able to take the real world and formulate it into a structure that can be solved through code and math. It’s also important to be a team player, have a pleasant personality, and be able to collaborate well with others. No matter how talented you are at coding, if you can’t work well with others, you won’t be a good fit.

So, it’s crucial to practice interviews and come across as a person who can function well in a team. You should be prepared for Star Method-type questions, such as questions about your working style, which are typically used as a filter to determine whether you’re organized and can work well with others. Remember, your ability to fit in with a team is just as important as your coding chops and data strategy.

Matt: Standing out as a data scientist in the current job market is not an easy feat, and there’s no secret formula for it. It’s a competitive field, and you have to be patient, persistent, and not get discouraged by rejection. While all the advice mentioned earlier is valid, we must also acknowledge the reality that many people are attracted to this field and are seeking these types of jobs.