Practicum Pride: TruSTAR
Viviana Márquez came to the MSDS program with a Bachelors in Mathematics and Neha Tevathia came to the MSDS program with a Bachelors of Engineering in Computer Science. Continue reading to learn more about their practicum experience at TruSTAR!
Can you tell us a bit about TruSTAR?
VIVIANA — TruSTAR is a threat intelligence platform that helps organizations leverage multiple sources of threat intelligence in real time to enrich investigations and support better and faster defensive actions. In their own words, “we’re obsessed with getting the most value from your threat intelligence data.” In order to do so, TruSTAR uses machine learning and automation to focus on building the best threat intelligence platform for enterprise cybersecurity.
Can you describe the projects you are working on?
NEHA — The current project we are working on tries to identify malicious text from documents using machine learning. We worked on exploring different methodologies to improve their current model and test the robustness of different models through various validation metrics and exploratory data analysis. The next project will involve natural language processing for malware detection.
VIVIANA — We are going to be working on three projects during our practicum. Our first project consisted of assessing the company’s malware detection model through different performance metrics and visualizations. We are evaluating whether or not their current algorithm is good enough, and if not, we’ll propose a new one. At first, I was both flattered and intimated by having such a big responsibility on my shoulders– this was not just another homework assignment. This was something that was going to get put in production for a company — but then as I started working on the project, I realized that our master’s program has given me the skills to rise up to the challenge.
How are you applying the knowledge gained from the program to your practicum? Is there a particular class that has been the most helpful?
NEHA — A lot of our knowledge gained during this practicum is based on the foundations we have learned in the program. The program’s courses on python and machine learning have been extremely valuable as it helps us build the critical skills required to execute real-world projects at TruSTAR.
VIVIANA — You know how machine learning algorithms’ performance are supposed to progressively improve with new data? Well, I feel like I am a ML model programming other ML models. I constantly find myself excited about the new things we are learning every day in class and wanting to apply them right away at TruSTAR. Due to the nature of our project, I would say that the Machine Learning class has been the most helpful so far, but it’s hard to say which class has been the most helpful because being a good Data Scientist involves having a wide range of skills from different fields.
What is it like working with professional data scientists?
NEHA — Working directly with professional data scientists helps us develop our machine learning skills and lets us directly apply what we are learning from the program. Not only do the data scientists help us understand the field better, but we also gain a lot of practical tips from their experience. As we know, machine learning is an applied science, hence it always helps to have someone experienced by your side mentoring you through the process.
VIVIANA — In the same way that some actresses move to Los Angeles to pursue their dream career in Hollywood, I moved to San Francisco to start a career in tech, and so far, my experience have lived up to my expectations. The environment at TruSTAR is very open, friendly, and creative. There are many whiteboards, computer screens, and specialty rooms available, making it exciting to work on projects and propose new ideas. In particular, our data science mentors in the company have been very kind to us by guiding us through projects, but also being genuinely enthusiastic and supportive about new ideas and skills we bring to the table.
Are there any cool perks?
NEHA — One of the cool perks is that we get to interact with other departments such as engineering and product, which will help us over the long term since most data scientists collaborate with these departments. Also, free food is always a valuable perk!
What is the biggest challenge you’ve faced in your practicum?
NEHA — A significant challenge we faced was understanding and applying machine learning techniques to text data, as generally text data is messy and unstructured. We have to double down on feature engineering and make sure we are able to decipher and transform the data before we finally use it.
For more information on the MS in Data Science program:
