I am a technology obsessed professional with a deep foundation in physics and materials science.
I currently lead high-impact AI and data initiatives as the Director of Data and AI at Tonal. I transitioned into a career in tech, software, and robotics after receiving my PhD in Physics and Materials Science from the University of California, Davis. Curious about the intersection of science and technology, I have used my research and acaemic background to advanced technological solutions for innovative companies in the Bay Area.
I'm always looking to connect with passionate individuals to discuss technology innovation, AI, data science, machine learning, robotics or sustainability.
Let's connectMy career has spanned across interests and industries, but what ties them all together is an insatiable curiosity and passion for building systems that make real change. After decades working to produce positive progress in science, technology, and sustainability, I've learned that the most productive way to make an impact is surround yourself with and collaborate eagerly with other curious people.
I obtained a summa cum laude M.Sc. in theoretical physics and a Ph.D in computational Chemistry. During my academic career, I was honored with a fellowship from the Molecular Sciences Software Institute under the NSF, where we developed the kALDo software—a cutting-edge tool for modeling heat transport at the nanoscale. My research has led to over ten publications in prestigious international peer-reviewed journals, including Nature Communications.
I am a data scientist with a background in software development and physics. I know how to write professional software and build mathematical models, and I have experience in machine learning, AI, and NLP. At Tonal, I develop AI models that process sensor data from hundreds of thousands of devices. Today I have five patents for AI and robotics, and my team has collected the largest strength training dataset in the world.
I have a deep interest in sustainability, specifically in material design. My academic research focused on designing efficient materials with high thermal performance. Additionally, I founded WeVux, a design blog that champions sustainable innovation. As part of that, I spearheaded the creation of a material design map, a tool that connects manufacturers and architects with sustainable materials, fostering eco-friendly practices across industries.
As Head of Data and AI at Tonal, I drive innovation at the crossroads of fitness, wellness, and digital health. By leveraging science, big data, and machine learning, we create intelligent, adaptive training programs that elevate performance and support long-term health and well-being. Our cutting-edge technology redefines the workout experience, making fitness more personalized, engaging, and impactful.
Discover Tonal
Heat has long been a complex and elusive concept, captivating scientists for centuries. Today, understanding heat transport is more critical than ever, especially at the nanoscale. Using advanced numerical models, we can simulate heat flow at the atomic level with remarkable precision. kALDo is a cutting-edge, Python-based software package that leverages both the Boltzmann Transport Equation and the Quasi-Harmonic Green-Kubo method, optimized for high-performance computing on GPUs and CPUs via TensorFlow. Released as an open-source tool, it empowers the scientific community to explore, innovate, and expand upon its capabilities.
Discover
kALDo
Past projects
During my time in Silicon Valley, I served as both an Engineering Manager and Software Engineer, collaborating with a diverse range of companies from startups to established enterprises. I led teams in developing and optimizing mobile and web platforms, helping businesses scale and enhance user experiences. This role exposed me to a variety of programming languages and frameworks, allowing me to deliver tailored, high-impact solutions in the fast-paced tech environment, while refining my leadership and technical skills.
Download my resume
We describe a theoretical and computational approach to calculate the vibrational, elastic, and thermal properties of materials from the low-temperature quantum regime to the high-temperature anharmonic regime. This approach is based on anharmonic lattice dynamics and the Boltzmann transport equation. It relies on second and third-order force...
AI-driven fitness technology is transforming how we approach active aging. Using real-time data from sensors, platforms like Tonal offer personalized strength training, improving longevity and health. By analyzing movement quality and providing tailored guidance, these systems enhance physical performance, helping users maintain functionality as they age...
AI-powered multi-agent systems are transforming how we interact with technology. Developers are increasingly implementing these systems to automate complex tasks and streamline user experiences. While single-agent applications are well-established, the full potential of multi-agent systems is only beginning to emerge as they prove their value in collaborative environments...
Recently, I worked on a patent involving the synthesis of exercise guidance data. This innovation generates modified exercise videos using a pose data change model, enhancing training data through varied labels. By training a guidance classifier model on these videos, the system improves the accuracy of exercise movement guidance, leading to more personalized and effective fitness recommendations...
I contributed to the development of a patent focused on Exercise Guidance Using Multi-Modal Data, which leverages advanced sensor technology to enhance workout efficiency. This innovation integrates data from optical and other hardware sensors to provide real-time movement guidance based on historical performance. By analyzing multiple inputs, the system triggers specific conditions to optimize...
At Tonal, I worked on a patent centered around exercise machine struggle detection. This technology predicts the performance of upcoming repetitions using data from prior movements, helping to detect potential physical failure during a workout. It determines how many reps a user has left in reserve, offering a tailored, efficient training experience. This advancement aims to improve workout safety and effectiveness...
Anomalous heat transport in one-dimensional nanostructures, such as nanotubes and nanowires, is a widely debated problem in condensed matter and statistical physics, with contradicting pieces of evidence from experiments and simulations. Using a comprehensive modeling approach, comprised of lattice dynamics and molecular dynamics simulations...
Understanding heat transport in semiconductors and insulators is of fundamental importance because of its technological impact in electronics and renewable energy harvesting and conversion. Anharmonic Lattice Dynamics provides a powerful framework for the description of heat transport at the nanoscale... Source code
The machine learning method has been applied for the first time to a binary compound. Researchers are increasingly using neural networks as a machine learning method to analyze the atomic structure of complex materials. Although the method has been applied to phase transitions in single compounds, its transferability across compositions of binary compounds has never been tested...
We introduce a novel approach to model heat transport in solids, based on the Green-Kubo theory of linear response. It naturally bridges the Boltzmann kinetic approach in crystals and the Allen-Feldman model in glasses, leveraging interatomic force constants and normal-mode linewidths computed at mechanical equilibrium...
We applied Vasiliki Plerou's model to the S&P 500 Value Index, analyzing demand-supply imbalance and price dynamics. The model, similar to an Ising model from statistical mechanics, highlights nonlinear price-change behavior. Our study explores how market imbalances correlate with price changes, revealing key insights into investor behavior and market...
In today's data-driven world, where everything is represented by probabilities, programming languages are evolving to include new native variables and distributions. Starting with an overview of probability theory and Bayes' theorem, I demonstrate how modern machine learning models can leverage probabilistic programming to make more accurate predictions and handle uncertainty in complex data environments...
In a world of data, where everything is represented by probabilities, programming languages are providing new kinds of native variables, distributions. Starting from an introduction of probability and Bayes theorem, in this video, we show how modern Machine Learning can take advantage of this new probabilistic programming paradigm... Source code
In 2011, the Canadian company D-Wave announced D-Wave One, the first commercially available quantum computer based on Rainer, a 128 superconducting flux qubits processor. Since very early age, the D-Wave computer had been used as a heuristic minimizer of the Ising energy functions. In the last few years, many institutions supported the research to use it as an optimizer...
Swift has gained a lot of popularity during recent years. From being a language used to develop only Apple devices, this versatile programming language has recently expanded its applications to many other fields. A recent example is Android development. In this video, we show how to quickly create a web server using Swift and IBM Kitura... Source code
With the acquisition of Next in 1997, a new tool was initiated into the Apple family. Originally known as an enhancement of OpenStep, called NextStep, it caught the attention of the developer community under the name of Interface Builder, as part of the XCode suite. Now about to celebrate its 20th birthday, Interface Builder... Source code
Graphic Processor Units are becoming more and more important in recent years and are spreading into many different fields, some of which include: computational finance, defense and intelligence, machine learning, fluid dynamics, structural mechanics, electronic biology, physics, chemistry, numerical analysis and security. There are many reasons why a... Source code
Several weeks ago, I set out to predict who would win the World Cup. I faced this interesting and challenging problem that I want to share: how can we accurately relate the outcome of the World Cup to the varying strengths, performances, and unpredictable factors influencing the teams throughout the entire tournament? Source code