My work spans end-to-end development of machine learning systems across both research and applied domains. I have experience designing and optimizing models for time-series forecasting, image classification, self-supervised segmentation, and Bayesian modeling. I’ve trained and fine-tuned transformer-based models for both vision and language tasks. My projects include building low-latency pipelines for embedded neurotechnology, developing quantification models for medical diagnostics, and applying reinforcement learning frameworks to simulate spatial cognition. I’ve also worked on biologically inspired benchmarks to evaluate large language models. I'm especially interested in building robust, interpretable ML systems that bridge scientific understanding and real-world impact.
🔬 Currently exploring:
• Chain-of-Thought monitoring
• Mechanistic Interpretability
• Neural computation and brain-inspired AI systems
• Models for scientific discovery
🛠️ PyTorch · TensorFlow · JAX · Python · Signal Processing