From thesis intern to a cyber security consultant at KPMG

Cristina is 23 years old and comes from Bucharest, Romania. She moved to the Netherlands almost five years ago to do her Bachelor in Computer Science and Engineering at TU/e. Cristina is now almost finishing her Master in Information Security Technology, also at TU/e. Meanwhile she is working on her master thesis for the KPMG Cyber Operations team.

Cristina: “As I was approaching the start of my master thesis, I started looking for companies which offered the possibility for doing my thesis there. Out of the ones I discovered, KPMG stood out with its flexibility in deciding what my topic would be, so I decided to apply. However, what convinced me to come here as an intern was the friendliness shown during both of my interviews and the help I got even before I started my thesis.”

Luc is currently following a Master in Cyber Security at the University of Twente and is now in the final phase of his master, namely doing his graduation research. For this, Luc had the choice to do his graduation at the UT or at a company. Since this offered a great opportunity to get to know a company from the inside and to experience it, he went looking for a suitable company. Luc was looking for a place where he had the freedom to do his research in the way he wanted, but also to be part of the team to see if he would like to start his career there.

Luc: “I was keen to get into consultancy and so quickly found myself looking at companies such as Deloitte, KPMG and Northwave. After talking to all three of them, I chose KPMG because I could decide for myself how I wanted to organize and conduct my research and because after the interviews I thought it would be the best place to work.”

Challenging assignments

For the graduation research of the Thesis Cristina is working on developing a framework for evaluating Extended Detection and Response (XDR) solutions, taking into account the bias of testing them in a small scaled laboratory.

Luc’s research is on Membership inference attacks in federated learning. Federated learning is a form of machine learning where you share a model with a network. Then everyone in the network trains their own copy of the model on the local data. Then the trained models are shared and merged into a new overall model. In this way, you can perform machine learning on large amounts of data that is spread across multiple places without having to share this data. This gives you more privacy and more computational power by training in parallel. But there is still a risk involved in sharing the model parameters. If these parameters are sent without encryption, someone can use them to retrieve information about the training data. This is called an inference attack and within that class of attacks Luc looks specifically at membership inference. With this attack you try to find out if certain data was used while training a model. For example, suppose you have data from a person named Mark. If he uses this data to attack a model trained on data from Alzheimer's patients then you can find out if Mark is in that dataset and has Alzheimer's, which is therefore a privacy or data leak. Luc’s specific question within this is what the effect is of different network sizes on the attack performance of membership inference.

Thesis versus projects

Cristina: “I did not get the chance to engage in other projects yet, but if I will be offered the chance, I would like to take up a challenge. However, I am observing a bit what the other members of my team are doing and I am planning to talk to them a bit about the work they do and what it involves.”

Luc: “My main goal that I am doing my graduation here at KPMG is mainly to do my research and write my thesis. But besides that, it is also important to get to know the company and the culture. There is a good balance now that I can decide for myself when I work on my research and when I join projects. It is also possible to join a project but I have not yet made use of that.”

Eagerness to help

Cristina: “The supervision has been great so far. Anytime I needed help with something I quickly got a response or it was taken care of quite fast. I consider that I am getting all the assistance I need and that everyone is eager to help.”

Luc: “The guidance is super. You get quick help with topics you run into and there is a lot of knowledge here that can help you directly with your thesis. Also, feedback on parts of your research or thesis is always on time and there is always room for a conversation about your graduation but also beyond that.”

Work, bonding and having fun

KPMG values a good work/life balance. Also, having fun with colleagues enriches the collaboration with the team and the way how you feel about work.

Cristina: “I believe being an intern here was one of the best opportunities I could have gotten. Besides the fact that everybody is eager to help me with my thesis when I need anything, everybody is very friendly and fun to work with. Even being an intern, I feel like I can talk to anyone (almost) like their equal, which is something you do not see everywhere. I believe this is really beneficial in a work place, as you are not afraid to express yourself and your ideas and indeed this is something I really appreciate. I have also participated in some activities with the team and with the cyber department and they have been really fun. I am really enjoying the fact that there are plenty organized activities which do not necessarily involve working, but rather bonding with each other and having fun.”

Luc: “The atmosphere is very good within the team. It feels to me like an extension of my student days. And that is really something I was looking for when I was looking for an internship. The feeling that you can go to anyone with your questions and there is always room for fun things to do besides work.”

Plans after internship

After the graduation of their master’s degree Cristina and Luc will continue their career at the KPMG family. Both will work as a consultant in the Cyber Operations team.

Cristina: “I just received my contract a few days ago and I decided to take the offer, as I would really like to continue working here.”

Luc: “I am already looking forward to gaining new experiences and knowledge.”