Engineering ideas into intelligent systems.

Working with state-of-the-art latent diffusion models toward the next generation of mobile models.

• Optimization and development of Stable Diffusion on native iOS devices with Apple Core ML.
• Investigation of memory and computational optimizations for large latent diffusion models.

I returned to TRIUMF to explore Bayesian optimization for tuning a particle accelerator.

• Used Bayesian optimization to improve tuning speed from four hours to ten minutes.
• Optimized and modeled complex systems with Gaussian processes.

I worked with a student-led engineering team to build an autonomous agricultural robot.

• Integrated onboard mechanical, electrical, and software systems for the autonomous robot.
• Managed a team of more than 60 students using Agile methodology.
• Brought the robot to the team's first competition at the 2023 METRICS ACRE campaign in Italy.

I applied machine-learning systems to real-world NFT fraud detection in a startup environment.

• Applied semantic segmentation to improve NFT fraud detection accuracy.
• Deployed self-supervised learning and vision-transformer models on AWS.
• Analyzed the latent embedding space for eight million images.

My time at TRIUMF developed my skills across software, research, reinforcement learning, and accelerator physics.

• Designed simulations for complex beamline physics.
• Applied policy-gradient reinforcement learning with LSTMs in a partially observed environment.
• Published as first author at the NeurIPS 2021 Workshop on Machine Learning and the Physical Sciences.
• Built object-oriented, multithreaded training pipelines for distributed multi-GPU training.
• Built a Python Tkinter user interface and integrated real-time data transfer with EPICS.

• Improved image-dataset preprocessing speed with custom Python and Bash scripts.
• Designed Docker environments for model training on cloud GPU clusters.
• Fine-tuned image-classification and object-detection models in TensorFlow and PyTorch.

• Supported a first-year physics course for engineering students.
• Prepared and led weekly tutorials on electromagnetism, circuits, and wave propagation.

• Supported a third-year machine-learning project course.
• Guided projects and taught computer vision, robotics control, and deep learning.

• Guided students through laboratory experiments with resistors, capacitors, and Arduinos.
• Prepared tutorials on data collection and analysis techniques.

• Designed a reflow oven and PCB boards for the competition vehicle's brake-light module.