Cyber Work: How data science and machine learning are affecting cybersecurity
In this episode of Infosec’s Cyber Work podcast, host Chris Sienko spoke with Anu Yamunan, VP of product management and research at Exabeam. They discussed her 18-year experience path of designing and securing products as well as how data science and machine learning are used in cybersecurity.
1. How did you first get interested in computers and security?
Anu first got interested in computers and security indirectly when she opened up to math, since computers and math are intertwined. Until the sixth grade, she was scared by math, but her teacher changed her perspective on the subject.
She saw her first computer in her junior year of high school, at which point she fell in love with computers and knew that computer science was what she wanted to study.
2. What are your daily job duties and favorite aspects of the job?
Anu leads product design, security and threat research at Exabeam. Her day-to-day can range from innovation and research about what is coming up in the cybersecurity space from a data science and machine learning perspective to enabling sales and talking to customers. She enjoys the breadth of her role and the fact that her job can differ so much from day to day.
3. What previous jobs and skills helped you springboard into your current position?
The first job Anu worked after college was as an engineer for HP, where she learned a lot. The hard part for her was transitioning from an engineer to a product manager. Being a product manager is a more macro role, where you need to understand market needs, the competitive landscape, your customers, etc. Anu dealt with this by looking for ways to go beyond what an engineer does to help out the product managers in any way possible. During this time, she earned her MBA.
Anu suggests you shed any fear you may have of wearing multiple hats, get out of your comfort zone, and embrace it. She learned this working for a startup for four years.
4. What are data science and machine learning?
From a high level, data science and machine learning are about algorithms that allow software applications to predict the future based on data. A real-world example of it is how Netflix can predict movie recommendations based on usage, age, sex and more. LinkedIn does the same for jobs and other things related to the professional world.
Data science and machine learning use large volumes of big data with these algorithms to create usage profiles and behaviors to predict future use and actions. At Exabeam, Anu uses this data-driven discipline to predict future cybersecurity threats and events. This is done by building fingerprints or profiles of users and devices within an environment to help predict threats.
The threats that Exabeam encounters most are cybersecurity threats and insider threats, both compromised and malicious (such as a former employee that steals intellectual property). One thing that can be said is data science and machine learning are disrupting the traditional SIEM space by enabling future forecasting.
5. What are the skills, certifications and background needed to get involved in cybersecurity machine learning and data science?
There are a variety of roles in the cybersecurity space that involve data science and machine learning. They range from product professionals (like Anu herself) who require strong analytical and data analysis skills to Chief Information Security Officers (CISO), which is more of a practitioner, and implementational consultants.
Anu suggests finding out what your specific cybersecurity passion is and focusing on it. Cybersecurity is the most under-employed space in the job market, which means there definitely is room for you.
6. What has been your experience as a woman in the cybersecurity field?
She does not notice it anymore after working in cybersecurity for 18 years, but she is definitely happy that there are more women in the field. While women are still underrepresented in cybersecurity, Anu says women have some innate abilities that make them particularly attuned to cybersecurity. Among them are a better ability to multi-task and attention to detail, as well as being more cautious and risk-averse. Women also offer a fresh perspective that the cybersecurity field can benefit from.
7. What are your thoughts on finding and recruiting more diverse candidates?
Exabeam has a lot of diversity across all positions in the organization, which includes a good number of women. From Anu’s perspective, it is not just about crafting a job posting to target women, but that women want to know they will be supported within an organization that promotes diversity. There is a marriage between the diverse perspectives and skills that women bring, which is what has brought Exabeam to where it is today.
8. What tips would you give to women entering the world of cybersecurity?
Don’t be afraid to use your voice as a woman because you get to have say too. Men are indeed more vocal than women and women do not like confrontation which compounds the issue. Anu’s take is that no one’s voice should not be heard. Be confident in what you know and what you are passionate about.
Exabeam uses monthly women’s lunches and women’s forums to make the organization more desirable to women. They have grown to involve over 50 people from different backgrounds and roles where they support, mentor, and learn from each other. This aspect of Exabeam is promoted at exabeam.com and LinkedIn and is attractive to women because of the emotional and social support it offers. They also use a referral program when hiring which has proven to be highly successful.
Work-life balance is huge for women and even Anu struggles with this. Having a strong support system at work and home helps out a lot in this regard.
9. What are some upcoming innovations in data science and machine learning that you’re looking forward to?
Where Exabeam came from five years ago, most of the systems were legacy and created before data science. This means they were not focused on predicting threats. About three years ago, they became less of a helper for SIEM and more of a competitor, with a platform that can predict threats by using what is called smart timeline to correlate user and device usage to predict future activity.
Exabeam is always innovating in terms of data science and focusing on data leaving the company. One of their innovations is identifying the personal email addresses of users because of insider threats. Spoofing is another area that is going to grow with the rise of machine learning because what may look normal to human eyes can be identified as a bot by machine learning.
10. Where can viewers find you online?
Anu can be found on LinkedIn and at exabeam.com. Exabeam is very active on social media, especially Twitter and Instagram.
In this episode of Cyber Work, Chris Sienko and Anu Yamunan spoke about how data science and machine learning are affecting cybersecurity. Anu brought valuable insight from a woman’s perspective, with a particular focus on the importance of diversity. Stay tuned for other episodes of Infosec’s Cyber Work podcast series!
To see the entire episode with Anu Yamunan, visit our YouTube channel.
- How Data Science and Machine Learning are Affecting Cybersecurity, Infosec (YouTube)