Could you tell us a little about yourself and your studies before coming to Imperial?
Before pursuing the MSc in AI, I completed a BSc in Neuroscience at UCL and an MSc in Translational Neuroscience at Imperial College London. During my neuroscience studies, I developed a strong interest in computational approaches, particularly advanced machine learning techniques for analysing neuroimaging data. As such, my MSc in Translational Neuroscience thesis focused on applying unsupervised statistical learning methods to study sleep fragmentation—essentially identifying different types of nocturnal awakenings by analysing EEG neurodynamics leading up to them.
What attracted you to the MSc in AI?
I had been interested in Machine Learning and AI since my final year of undergraduate studies. While I made efforts to self-study, take relevant modules in machine learning and statistics, and apply AI concepts in my work, I realised I lacked a strong formal foundation — particularly in mathematics, coding, and the theoretical principles behind machine learning. This gap became especially apparent while working on my first MSc thesis, where I found myself limited in how deeply I could integrate AI-driven methods and analysis.
At the same time, I was fascinated by the parallels between neuroscience and AI: the way deep learning drew inspiration from neural structures and neurophysiology, or how reinforcement learning echoed principles from psychology and cognitive science. It made me curious about other unexplored synergies between these fields, particularly in areas like neurodynamics, information theory, and control theory applied to brain-computer interfaces and non-invasive brain stimulation paradigms.
That’s what motivated me to pursue further studies in AI. When I came across Imperial’s MSc in AI, it felt like the perfect fit — offering a rigorous, in-depth immersion into AI fundamentals, mathematical and statistical foundations, and hands-on coding experience, all topped off with an independent research project. I applied as soon as I could and was thrilled to receive an offer.
What did you enjoy the most?
One of the biggest highlights was the numerous pieces of coding coursework, and projects. At first, they were quite intimidating: having never done AI-focused coding at such a level, I genuinely doubted whether I could manage. But the challenge led to a steep learning curve, and looking back, I’m amazed by how much progress both my classmates and I made.
Going from building a simple decision tree to estimate house prices based on property details to developing a web app for real-time prediction of biophysical properties of drug molecules from their structure using attention-based and convolutional graph neural networks — all in just four months — still feels unreal when I reflect on it. These projects provided hands-on experience across different areas of AI and proved invaluable when applying for PhD positions.
Another aspect I loved was the final research project. The MSc allows students to propose their own topics, and I took the opportunity to continue my research into sleep neurodynamics, this time equipped with AI-driven time-series analysis techniques. It was incredibly rewarding to explore a topic I was passionate about with the tools I had developed throughout the course.
Finally, the social environment was fantastic. Our cohort came from incredibly diverse backgrounds: some straight out of their undergraduate degrees, others transitioning from entirely different careers. Despite the challenges of the course, there was a strong sense of friendship, support, and learning from each other, and I formed lasting friendships that made the experience even more rewarding.
What did you find most challenging?
The sheer pace of the course, combined with the depth and volume of material, was intense, especially since I started with a limited background in mathematics and statistics for AI and had mostly coded in MATLAB rather than Python. The workload was demanding, but it pushed me to develop much better time-management and learning strategies.
Understanding the material from different angles, rather than just memorising it, was crucial for the exams, and that was a big shift in how I approached studying in the state of information overflow. Looking back, I’m still in awe of how much I learned in just one year.
Could you tell us about some of your achievements on the MSc that make you proud?
One of the things I’m most proud of is my personal growth throughout the course. If you look at my grades, you can clearly reconstruct my learning curve — it’s almost perfectly correlated! By the second term, I had developed strong skills, ultimately excelling in my exams (with an average in the 80s) and research project (earning a high distinction) and finishing my MSc with a strong overall distinction.
There were definitely some moments of self-doubt and a few long nights in the library. But thanks to the encouragement of our MSc AI coordinator, Dr Rob Craven, and the support of my peers, I felt empowered to push through the challenges and reach a level I wouldn’t have thought possible at the start of the year.
What did you do in your spare time?
I made an effort to stay active whenever possible — going for runs, hitting the gym, and attending dance classes. The taught component of the MSc was comparatively busier, but once the research project started, I found it much easier to balance exercise with work. Staying physically active really helped keep my mind sharp and focused.
Could you tell us about your individual project?
My research project was an interdisciplinary collaboration between the Department of Computing and the Department of Brain Sciences at Imperial, supervised by Dr Pedro Mediano in collaboration with Prof. Nir Grossman and Dr Junheng Li.
The project focused on using deep learning and advanced machine learning techniques to assess proximity to sleep and predict sleep onset events in real time using EEG data. I had significant creative freedom while also receiving invaluable guidance from my supervisors. Throughout the project, I experimented with various architectures for the predictive pipeline, including Transformers, LSTMs, Echo State Networks (ESNs), Random Forest, XGBoost, and LightGBM. By the end of the project, the results were exciting — we were able to predict sleep onset, on average, 10 minutes before it occurred using EEG signals from just three EEG channels.
What have you been doing since you graduated?
After graduating, I worked as a Research Assistant in Prof Nir Grossman’s lab for a year, contributing to a publication. I knew I wanted to pursue a PhD, so alongside my research work, I applied to various programs. I received multiple offers from both Cambridge and Imperial but ultimately chose to stay at Imperial. I was also awarded a Global Talent Visa, which was a great recognition of my work.
Since September 2024, I have been undertaking a fully funded PhD in Clinical Medicine at the Brain Sciences Department, with Dr Cynthia Sandor as my primary supervisor and Dr Pedro Mediano and Prof. Payam Barnaghi as my co-supervisors. My research focuses on applying AI and causal inference methods to the early detection of Parkinson’s disease and drug repurposing using digital medical data, such as electronic health records, smartwatch data, and EEG/ECG signals.
I don’t think I would be where I am today without the growth and experience, I gained through the MSc, and to this day, I am grateful for the skills, knowledge, and professional connections it has fostered.
Do you have any advice for prospective students?
Expect to be busy! The taught component of the MSc is intense, so strong time-management skills are crucial. It’s a challenging program but also incredibly rewarding. So, don’t be too hard on yourself if things feel difficult at first. A growth mindset is essential!
Collaborate with your coursemates and build friendships — it makes everything easier and more enjoyable. For exams, start practising problem-solving early, as understanding the concepts, and getting hands-on practise in solving problems matters far more than rote memorisation.
If possible, try to complete the pre-course materials before the academic year starts, they really help set you up for success. And most importantly, try to enjoy the experience. The year flies by, and despite the challenges, you’ll likely look back on it as one of the most transformative and fulfilling periods of your life.