I am an eternal student of Artificial Intelligence with a deep-rooted passion for the hardcore mathematics that power foundational models. My primary obsession lies at the intersection of AI and Computational Biology, where I apply advanced machine learning techniques to accelerate drug discovery and enzyme engineering.

Beyond the application layer, I am deeply fascinated by the theoretical foundations of deep learning. In my free time, I enjoy exploring differential geometry, particularly tensor algebra, Riemannian manifolds, and quantum operators. Currently, I am moving down the stack to master hardware-level execution, writing custom kernels to understand the exact mechanics of GPU compute during training and inference.

• 🔭 I’m currently working on: Developing MolFun, an open-source framework designed for finetuning structural and sequence models like OpenFold, Boltz, various biological embeddings, and de novo generation models such as RFdiffusion. The project is built to seamlessly test novel architectures and run comprehensive benchmarking.
• 🌱 I’m currently learning: Low-level GPU architecture, custom kernel development (CUDA/Triton), and the intricate mechanics of hardware optimization for AI training and inference.
• 👯 I’m looking to collaborate on: Open-source projects related to AI in biotech, structure-based drug design, foundational models for biology, or High-Performance Computing (HPC) optimizations.
• 💬 Ask me about: Tensor algebra, Riemannian geometry, the mathematical proofs behind neural networks, and how AI is revolutionizing protein engineering.
• 📫 How to reach me: rubencr14@gmail.com | LinkedIn
• ⚡ Fun fact: I consider discussing Riemannian manifolds and parallel transport a perfectly valid and relaxing hobby.

• Domains: Computational Biology, Drug Discovery, Protein Engineering, Deep Learning.
• Mathematics: Linear & Tensor Algebra, Differential Geometry, Calculus.
• Engineering: PyTorch, CUDA, Triton, GPU Optimization, HPC.