Amsterdam, North Holland, Netherlands
As a data scientist who works with astronomical data (galaxy clusters), I developed Python algorithms to interpret, described, and model observations using Bayesian statistics. I am looking forward to working in a fast pace environment, where the impact of the work and the results obtained are tangible. I have lived in 3 different continents (Asia, America, Europe), and I have always worked and studied in a diverse environment. This has helped me become a good leader and team member when needed who adapts fast and easily to workplaces and who can communicate effectively. I also love the thrill of discovering and learning the customs of a new place. I have 5 years of experience in research and have work on very different projects in nature. During my undergraduate studies, I gained experience in data handling and reduction. From my master's and Ph.D. research, I have developed abilities in statistics, data analysis, and problem-solving. As a personal interest, I have always been involved in outreach activities such as organizing science talks for the general public as well as teaching science to kids.
Lead the design and cross-functional integration of data-driven solutions, collaborating with marketing, sales, and program management to identify and test high-impact use cases. Oversee development and validation of machine learning models (e.g., ETA prediction, clustering) to enhance customer supply chain performance; mentor junior scientists throughout modeling and accuracy evaluation. Initiated and have continuously led biannual innovation sprints to accelerate proof-of-concept development and advance team capabilities in applied data science; conceived the strategy, secured leadership support, and fully designed its implementation. Built on prior collaboration with academic and industry partners to take a leading role in the refinement and deployment of digital twin simulations using time series data; guide the experimental design and validation of new models as an industry expert to ensure real-world applicability and impact on cold chain logistics
Working as the main data scientist, leading innovation projects to utilize the information provided by Sensitech data loggers to create predictive tools to improve the cold chain process and the quality of the products. As well as overseeing the full integration of the developed tools into the system, including internal POCs, POCs with clients, and full implementation. Therefore, working with different stakeholders to help carry an innovation project to production. Projects include: *Development of tools to identify temperature excursions and quantify their effect on the product *Evaluation of produce quality after a cooling process *Development of process control analysis and dashboard based on historical data per client.
In terms of Astronomy, I estimate the outer slope of the universal pressure profile (gNFW) for a sample of galaxy clusters using MCMC. The galaxy clusters were observed with the Sunyaev Zeldovich effect in sub-mm wavelengths. The high pass filter applied to the data (in the reduction steps) was modified to have more signal in the outer regions of the clusters/ images, allowing us to measure the desired parameters. The purpose of such work is to improve the detection templates used to discovered new clusters when observing over large areas of the sky. The detection templates tend to use a generalized profile. However, if the clusters have been disturbed the outer slope of the pressure profile could be different from the one for the relax clusters. My work aims at obtaining a general parameter estimate for the outer slope of the pressure profile of these two sub-groups. Having pressure profiles that take into account the dynamical state of the clusters, will allow us to detect more clusters and therefore improve on the precision of cosmological models.
Master thesis research work carried under the supervision of Prof. Alberto Franceschini. Study of the two-point correlation function between galaxies using infrared observations. I developed python codes that use statistical tools to determine the distribution of sources and the degree of clustering between them. The purpose of the thesis was to determine the clustering factor between the galaxies in a determined field.
Determining variable star properties in globular clusters: - Combining archival data taken over 10 years with different instruments. - Development of data visualization tools in Perl and Supermongo.