Santa Cruz, California, United States
I'm a machine learning and data scientist with a deep-rooted passion for data analytics and high performance computation, backed by over a decade of experience in teaching, consulting, and research. I have two masters degrees (Physics and Statistics) and a PhD in risk analysis in radionuclide geological disposal sites. My expertise spans Bayesian statistics, functional data analysis, and neural networks, with a career history that includes pioneering work at Research and Development in Aerospace for the European Space Agency and high-performance computing consultancy at Cognizant Mobility. I excel in leveraging mathematical tools for problem-solving and have a strong track record in applying statistical models to a variety of fields—ranging from aerospace technology to nuclear waste disposal. As an educator, I've co-authored textbooks, mentored junior teachers and led courses in advanced statistical linear models, linear algebra, game theory, and differential equations, always with a focus on elucidating the 'why' behind the 'how.' My skill set includes predictive analytics, deep learning, and high-performance parallel computing, among others. If you're looking for someone who thrives in tackling complex problems and translating them into actionable insights, I'm eager to explore new opportunities in data science and machine learning. Programming skills: Tensorflow, Python, SQL, R, SAS, Matlab, Mathematica, C, C++, Fortran, Bash. Mathematics/Statistics: Convolutional Neural Networks, Hopfield Neuronal Networks, Bayesian Statistics (Markov Chain Monte Carlo simulation), hierarchical modeling, linear models, Functional Data Analysis. Applied Mathematics (Linear algebra, diferential equations, modeling of physical systems).
Developed deep learning and data analytic tools for detecting, localizing and mitigating radiofrequency interference in satellite imaging.
Implemented real-time detection of image features via GPU processing and CUDA programming.
We used Reversible Jump Markov Chain Monte Carlo simulations to infer change-points in contaminant dispersion data. We used fiarly data obtained from field experiments that produced data on how nuclear contaminants disperse in the geosphere; we were able to infer at least two change-points in the data (somehting that becomes obvious only after we found them).
Developed a set of e-classes in Statistics. The company was developing a set of lessons on Introductory Statistics to be delivered exclusively on an iPad. I also supported the development of statistical simulation tools that went along with the set of lectures.