Greetings! I'm Eric, currently serving as a Senior Data Scientist at Novartis Digital Finance Hub.
My educational background includes a major in Biochemistry, a master's degree in Bioinformatics and Biostatistics, and various certifications in areas like Deep Learning, AWS, and Industrialization of data-driven products. My expertise lies in harnessing the power of AI to tackle real business challenges, enhancing efficiency, precision, and execution speed while reducing costs for companies.
In my entrepreneurial journey, I've spearheaded diverse data-driven projects spanning multiple domains:
A - Developed a computer vision solution for skin cancer detection (http://biop.ai/).
B - Collaborated with onco-nutritionists, employing LLM models to create a comprehensive database of diets addressing treatment-related side-effects in cancer patients (https://nutrilieve.com/).
C - Engineered analytics and price tracking tools for the Spanish real estate market (https://www.estate-tracker.com/).
I am deeply passionate about leveraging AI to drive impactful solutions and enhance business outcomes.
OpenSea (https://opensea.io/es) is the biggest NFT marketplace. In such a volatile market, keeping track of the floor price of the NFT collections is extremely important.
The aim of this project is to automate the the tracking of floor price of any NFT collection and trigger mailing alerts when those collections reach a particular price.
Technologies: Python, Google Actions
This repository was developed with the aim to automatize the preprocessing, training & testing of multiple models that can be leveraged in Datathons or Kaggle Challenges.
The in-built classes can easily work with AWS technologies to automate Hyperparameter tunning workflows to fasten the development of products in such quick-challenges.
Technologies: Python
Sentiment analysis on stock news, press releases & earnings call has been widely used with the aim to capture positive sentiment or financial KPIs that could drive a positive impact on the stocks.
In this project authors leverage ChatGPT, a Large-Language Model [LLM], with the aim to automatically obtain insights on CFO interventions for the different stocks in S&P500.
Technologies: Python
Leverage a Streamlit app & the U2NET architecture in order to scale & remove the background of any image of intereset.
The architecture automatically identifies the items in the image and provides a grey-scale mask that can be used to remove the background from the original image.
Technologies: Python & Streamlit
VeriNA3d is an R package for the analysis of Nucleic Acid structural data. The software was developed on top of bio3d (Grant et al, 2006) with a higher level of abstraction. In addition of single-structure analyses, veriNA3d also implements pipelines to handle whole datasets of mmCIF/PDB structures.
As far as we know, no similar software has been previously distributed, thus it aims to fill a gap in the data mining pipelines of PDB structural data analyses.
Contributed through enhancing different features for a faster processing of the data, calculus of dihedral angles, among others.
Technologies: R
This repository contains the R package for Epidemiological Foci Relating Infections by Distance (EpiFRIenDs), a software to detect and analyse foci (clusters, outbreaks or hotspots) of infections from a given disease.
Contributed through enhancing multiple features of the core algorithm & implementing an application deployed using RShiny technology, available in Release 2.0 of the software.
Technologies: R, RShiny
