👋 Hi, I'm Tristan.
I’m a research scientist and machine learning engineer with experience in both AI and computational neuroscience, bridging theoretical insight and practical implementation.
- Novel approaches in machine learning for real-world impact
- AI safety and ethics
- Neuro-symbolic systems
- Computational Neuroscience
- Data analysis and interpretability
EEG Seizure Detection System (2024-Current) Developing a vector-quantized variational autoencoder based diagnostic system to assist neurologists in classifying seizure types from EEG brain signals. The architecture uses a novel Variational Autoencoder tokenizer that applies Fourier transforms to raw time series data, enabling the transformer to process neural oscillations similar to how LLMs process language tokens. This approach bridges signal processing with modern NLP architectures, creating interpretable representations of brain states that could revolutionize clinical EEG analysis.
Abstract Reasoning Corpus (ARC) Challenge (2024-Current) Leading development of a novel neuro-symbolic approach to tackle one of AI's most challenging reasoning benchmarks. The ARC challenge tests visual reasoning abilities that current AI systems struggle with, requiring the integration of pattern recognition, abstract thinking, and systematic generalization. My methodology combines neural pattern detection with symbolic rule extraction, aiming to achieve human-level performance on tasks that require genuine understanding rather than memorization.
Connectome Harmonics: Brain Dynamics Across the Lifespan (2022-Current) Directing an international research collaboration analyzing how structural brain changes influence neural dynamics throughout aging. Using a novel mathematical framework called "Connectome Harmonics," we're processing large-scale neuroimaging datasets from the University of Oxford to reveal new insights into cognitive decline. Our findings suggest potential therapeutic targets and early biomarkers for age-related neurodegeneration. This work combines spectral graph theory, complex systems analysis, and large-scale data processing.
I draw inspiration from thinkers like Joscha Bach, Douglas Hofstadter, Selen Atasoy, and Leonid Perlovsky.
Outside of research, I explore meditation, consciousness, philosophy, and music. I've traveled to 16 countries and counting.
This GitHub showcases a subset of my work. Some projects are proprietary, unpublished, or in progress—but the public repositories here reflect areas of active research and technical interest.
Feel free to reach out with an opportunity or colaboration idea! Contact me via email or connect on LinkedIn.