Jose A Dianes

I came to R by chance, while working in a bioinformatics institute. There, we had to provide tools for researchers to deal with biological data. I have to confess that at first I didn't like it. Being a software engineer I prefer languages that allow you to build and scale complex software systems. R is not meant to be that way. The language is build for statistics and data analysis work, and has probably the richest ecosystem for that purpose. In the end I loved to work in RStudio and using DataFrames, ggplot2, and build all sort of machine learning models with just a few lines of code. Very powerful tool that you can now even combine with Apache Spark!

Data visualisation is a natural and powerfull way to show numbers and stories behind data. I normally start by doing some Exploratory Data Analysis work. Then, once I find the story I want to tell, it will depend on what the intended audience is. For single image reports, I normally take a naked chart into some design tool like Sketch and work around it. Alternativelly, I can use R markdown to generate a full data report or go all the way to Shiny, possible powered up with JavaScript visualisation libraries such as D3.js.

Python, the most versatile language. But it is in machine learning and AI where I enjoy it the most. Powerful libraries and tools. Clean code. The right choice most of the time.

I came to machine learning and data analysis from a Software Engineering background. I have a PhD in real-time distributed systems and my interest is to build data products that give companies the edge. I am particularly interested in building data products that can scale to large amounts of data and work in real-time.

I came to Spark from a software engineering background. We had a Hadoop pipeline for clustering proteomics spectral data. Lately I've been working on on-line spectral search and on several Spark tutorials.

A fundamental part of data analysis, machine learning, and research, is being able to do statistical inference and understand the mathematics behind the computations. Coming from a computer programming background, I had to apply myself intensively in order to understand why and how to use the methods in the data scientist toolbox.

While working as a computer scientist I've collaborated with biologists and bioinformaticians in facilitating and improving their analyses. I've used different R and Python packages, together with big data and cloud technologies.