Website • Email • LinkedIn • ORCID • Google Scholar

I'm a PhD student in Electrical & Computer Engineering at the University of Florida, building machine learning for physical and neural systems. My current research learns surrogate models for computational fluid dynamics — fast stand-ins for expensive physics solvers, held accountable to the physics they approximate. That sits on a foundation of signal processing, time-series ML, and neuroengineering.

• Researcher, SmartDATA Lab, University of Florida — CFD surrogate modeling with Dr. Joel B. Harley
• Researcher, Computational NeuroEngineering Lab (CNEL), University of Florida
• President (2026–2027) & co-founder, IEEE Signal Processing Society @ UF
• Interests: neural operators, reduced-order modeling, state-space models, online/streaming learning, interpretable ML under noise

• CFD Surrogate Modeling – Learned surrogates for computational fluid dynamics: neural operators, mesh graph networks, and physics-informed training, judged on rollout stability and physical consistency.
• Kernel Adaptive Memory – Preregistered experiments on whether kernel-routed persistent memory helps sequence models.
• Zebrafish Voltage-Imaging & Neuromechanics Analysis – Event detection and source separation for optical voltage imaging.
• SPS Curriculum – Standardized, hands-on workshops for signal processing & ML at UF.
• Plato’s Cave – A human-centered system for validating research papers (arXiv:2603.23526, contributor).

I care about robustness, latency, and interpretability: if a method is fast and reliable, we can use it in the loop with humans.

Languages: Python, C/C++, (some) Assembly, TypeScript Compute: CUDA, CuPy, PyTorch, RAPIDS, Numba DSP/ML: Surrogate modeling, neural operators, signal processing pipelines, state‑space models, Kalman filters, online learning Data/Apps: Jupyter, FastAPI, Next.js/React, Discord API, OpenAI API Workflow: Git, GitHub Actions, Conda, VS Code

• SPS_Curriculum · Jupyter workshops for SP/ML at UF → https://github.com/Jibby2k1/SPS_Curriculum
• Zebrafish‑VoltageImaging‑Neuromechanics‑Analysis · Python analysis utilities for optical voltage imaging → https://github.com/Jibby2k1/Zebrafish-VoltageImaging-Neuromechanics-Analysis
• DinoDetect · Discord analytics + LLM‑assisted trend discovery → https://github.com/Jibby2k1/DinoDetect
• copit · Mobile, recommendation‑style shopping prototype (TypeScript) → https://github.com/Jibby2k1/copit
• PythonDJ · Small DSP + music tooling experiments → https://github.com/Jibby2k1/PythonDJ
• SwampScheduler (fork) · Scheduling app experiments → https://github.com/Jibby2k1/SwampScheduler
• WorkshopArchive (fork) · Archived workshop materials → https://github.com/Jibby2k1/WorkshopArchive

Full project write-ups, with diagrams and method notes, live at raulv.dev.

• Numerical linear algebra & modern analysis (for ML theory)
• Online & adaptive filtering for real‑time inference
• OS / systems for reproducible, low‑latency pipelines

If you’re building scientific ML, assistive neurotech, robust DSP/ML, or open education resources, I’m happy to chat. Best way to reach me is by email or LinkedIn.

When I’m not coding, I DJ and produce electronic music (dance, trance, techno) and run hands‑on workshops for the UF community.

Thanks for stopping by—enjoy the repos and say hi!