Suggestions
Leandro Ferrado
Data Science & Machine Learning
Leandro Ferrado is a Data Scientist and Machine Learning Engineer currently working at CloudX | Nearshore Software Development.2 He is known for his expertise in data science and machine learning, with a background in informatics engineering from FICH-UNL (Facultad de Ingeniería y Ciencias Hídricas - Universidad Nacional del Litoral).3
Professional Background
Ferrado is passionate about his field and has demonstrated his knowledge through various platforms:
- He has given talks on data science topics, including a presentation at PyData on "Breaking the 80/20 dilemma in Data Science projects".1
- His GitHub profile showcases his work as an Informatics Engineer focused on Data Science & Machine Learning.3
Expertise and Interests
Leandro Ferrado's areas of expertise include:
- Data Science: He has shared insights on data team structures and best practices in the field.4
- Machine Learning: His background combines both data science and machine learning engineering.2
- Project Management: Ferrado has discussed optimizing efforts in data science projects, particularly addressing the "80/20 dilemma".1
Online Presence
Ferrado maintains an active online presence:
- His LinkedIn username is leferrad, where he shares professional insights and experiences.24
- He has a GitHub account under the same username (leferrad), where he likely shares code and collaborates on projects.3
Leandro Ferrado appears to be a well-rounded professional in the data science field, combining technical expertise with practical insights on project management and team structures in the data science industry.
Highlights
Very good article about the way to structure your Data Science team. In my perspective, the easiest approach is to have "embedded" teams that you can allocate to business units based on the needs (and this will scale very cheap). But it is also the riskiest one since the members of your team will also "be embedded" in the ways of the business units (e.g. execution, know-how, tempo) to the point of not feeling like an actual team working side by side for the same goals, and then the engagement of your team will be strongly affected (trust me, I know about this). Therefore, I usually prefer having a "hybrid" approach, at least in terms of managing cells based on the main roles (e.g. data engineering cell, data science cell, backend cell). The best practices of the roles are reinforced in the cells, and alignment is achieved between cells to pursue the same strategy. The management required will be higher, but at the end you will have a higher benefit with teams that work aligned in terms of practices and shared knowledge that makes easier to take new Data Science challenges.
After many great talks with different colleagues and some good reading, I've finally decided to put in words some thoughts about this very interesting topic of Developer Experience (DX), in terms of what I've been experiencing in this fantastic world of Data Science. #datascience #dx
