About
I'm a machine learning engineer and data scientist with education in mathematical engineering and economics, interested in the application of mathematical and statistical models for solving real-life problems. I have experience working as a data scientist with financial models and as a consultant for the banking industry. I'm a very curious and responsible person who always wants to see the world in new ways and try different tools to solve problems. I have been trained to approach problem-solving in a systematic and organized way, applying quantitative models to advice companies to take better decisions.
ML Engineer & Data Scientist
- Phone: +57 320-333-8151
- City: Medellín, Colombia
- Degree: Professional
- Email: otalorajuand@gmail.com
Skills
My expertise lies in crafting specialized mathematical and statistical model tools, primarily using Python. My toolkit extends to encompass a diverse range of applications, while I also possess a solid foundation in software development with languages such as JavaScript. Here's an overview of the skills I bring to the table:
Resume
Sumary
Juan David Otálora
Creative mathematical engineer and economist with 3 years of experience in mathematical models, machine learning, data science and data engineer.
Education
B.Sc. Economics
July 2013 – August 2019
EAFIT University. Medellín, COL
As an economist, I could understand how to see the world through a model, how to simplify difficult and complex problems with mathematical equations and obtain worthy predicitons. I specialized in econometrics and quantitative economics.
B.Sc. Mathematical Engineering
January 2015 – December 2019
EAFIT University. Medellín, COL
With the help of the mathematical engineer bachelor, I could learn how to approach complex problems with a wide toolkit of methods and to be flexible in undestanding real-life situations from different optics. I specialized in statistical inference and ML models.
Software Development Bootcamp
June 2022 – February 2023
Holberton School. San Franscisco, USA (remote)
In the Holberton Bootcamp I developed foundations for software development and computation, I learned low level languages like C and Bash and high level languages like Python and JavaScript.
ML Bootcamp Specialization
March 2023 – December 2023
Holberton School. San Franscisco, USA (remote)
I specialized in Machine Learning and Deep learning and learned how to implement the cutting-edge models of this area.
Professional Experience
ML Engineer
Nov 2023 - January 2024
Casa Museo El Santuario - El Santuario, COL
- Developed a Chatbot based on AI that centralized the information of the institution and the history of the town of El Santuario in Colombia.
- Collected information from the town's history and administrative procedures to create a database for the institution.
- TECH-STACK:Python, HuggingFace, OpenAI, Streamlit
Backend Software Developer
Nov - Dic 2022
Techstars - Miami, USA (remote)
- Developed a centralized web app for Techstars, streamlining client data and contact search.
- Design and implement a database for the clients to improve the data flowing of the company.
- TECH-STACK: React, FastAPI, MySQL, Pandas, NodeJS, SQLAlchemy
Data Scientist
Mar 2020 - Apr 2022
Dataknow - Medellín, Col
- Developed clustering models for client segmentation using RFM, and DBSCAN models for a cosmetics company as a consulting project in order to have a better understanding of their clients and adjust their marketing campaign.
- Worked for the Colombian bank Davivienda in the IFRS9 financial area to support and develop the data ETL to feed the risk models. I also supported the development of the rules for the categorization of risk clients.
- Due to a partnership with the Colombian bank Bancolombia in the client's quality center, I developed ETLs for the data flow of this area. I worked with high-volume data.
- Designed PowerBI dashboards for a Saas initiative that allowed companies to observe their clients in a segmented optic and keep track of their most KPIs.
- TECH-STACK: SAS MIP, SAS Data Integration, SAS Guide, Hadoop, Impala, SQL, Excel, PowerBI
Data Analyst Intern
Jan - Jun 2018
Protección - Medellín, COL
- Supported and maintained the quantitative models for the analysis of the portfolios of the investment area.
- Implemented the translation of the automation tasks for rebalancing and updating the data of the portfolios from Visual Basic to modern programming languages (R and Python).
- Developed and maintained an R Shiny app for the econometrics models of the specialists, as the internship final project.
- TECH-STACK: Visual Basic, Excel, R, Python, R Shiny
Courses
- Programming Complete; Learn SAS and Become a Data Ninja - Coursera (2020)
- Machine Learning and Data Science with Python - Coursera (2020)
Portfolio
These are the programming projects I have worked on. I have worked on low level projects with languages like C and Bash, where I created a command line interpreter. I also worked on high-level languages like Python and JavaScript on back and front end web development and SQL databases. With these technologies I developed an AirBnB clone and I worked on a software development project for Techstars. I have also worked on ML projects, developing the most used Deep Learning models for a deeper understanding and i also have a personal projects repo where I delevop the ideas that come to my mind for applying the ML concepts I learn on the way.
- All
- Learning Repos
- Final Projects
- Personal
Academic Projects
I have had the opportunity of working in academic projects that added value to my formation. I mostly work in projects related to portfolio optimization with alternative techniques to estimate it, worked with functional data estimating the third and foruth moment of this type of data, and I also worked in the Center for Economics Research of my University as an research assistant on microeconomics econonometrical models.
Robust and Non parametric Portfolio Selection (2018)
Professor Henry Laniado - EAFIT University
A research practice based on a method for generating robust and non parametric portfolios for the Markowitz model.
Skewness and Kurtosis Based on Quantiles for Functional Data (2019)
Professor Francisco Zuluaga - EAFIT University
A research practice for translating the concepts of Skewness and Kurtosis based on quantiles for a descriptive analysis of functional databases.