Martín Levy Daniel
Verified Expert in Engineering
Statistics Developer
Buenos Aires, Argentina
Toptal member since April 9, 2021
Martín is a versatile data scientist interested in challenging projects. He has worked on numerous projects throughout his professional career in different fields such as finance, market research, technology, and, currently, in GIS and emergency management. Martín stands out for his high adaptability, problem-solving skills, and communication of solutions in business environments.
Portfolio
Experience
- Finance - 4 years
- Statistics - 4 years
- Python - 4 years
- Machine Learning - 4 years
- PostgreSQL - 3 years
- Forecasting - 2 years
- GIS - 1 year
- PyTorch - 1 year
Preferred Environment
PyCharm, Slack, Jupyter Notebook, Git, Visual Studio Code (VS Code), AWS IoT, Google Cloud Platform (GCP)
The most amazing...
...thing I've done was train a deep learning architecture to predict building roof type using LiDAR data to assess resilience in the event of a natural disaster.
Work Experience
Senior Data Scientist
Dorothy
- Developed a machine learning model that predicts flood risk better than the maps provided by FEMA by using updated maps as input. This flood model assesses actual flood risk more accurately, improving the F1 score by 40%.
- Developed a credit risk model from scratch for the loan product using reference data scraped from online lending platforms. This helps us benchmark how we should charge our users and identify the most important attributes to use.
- Developed a statistical model to estimate the probability of claim payment to identify eligible users for a loan product.
- Worked on a monitoring system to identify regions severely affected by weather events in order to carry out our marketing campaigns and acquire new users.
- Created the whole data pipeline for processing and loading GIS data into a Postgis database in AWS virtual machines together with data engineers.
- Trained and deployed computer vision models with PyTorch to assess building characteristics using remote sensing data such as LiDAR and high-resolution satellite imagery.
Product Data Scientist
OLX Global
- Developed a complete forecasting package in Python to assess the performance of product releases and marketing campaigns. This package included many models and was used as a benchmark for many migrations and marketing campaigns.
- Developed a lifetime value analysis using survival regression analysis. This model identifies factors that contribute to users leaving the platform, identifies major product pain points, and predicts which users are most likely to leave.
- Analyzed and A/B tested the release of new recommender system models to measure their performance and gain a complete understanding of how users search for items. Developed product roadmap together with product managers based on results.
Market Research Analyst
Market Platform Dynamics
- Headed an overall market analysis to assess how users dealt with the checkout process for PayPal and Visa. This involved collecting online and physical survey data and acquiring databases to understand the checkout performance.
- Automated and improved data processing and analysis. Through Python and the development of a data warehouse and data processing, significantly sped up and reduced errors.
- Developed inferential statistical analysis to measure the impact of different attributes on customers' products. This involves the use of linear models and causal inference techniques.
Experience
Forecasting Tool
Lifetime Value Regression Analysis
Education
Postgraduate Degree in Data Science
Universidad de Buenos Aires - Buenos Aires, Argentina
Master's Degree in Econometrics
Universidad Torcuato Di Tella - Buenos Aires, Argentina
Licenciate Degree in Economics
Universidad Torcuato Di Tella - Buenos Aires, Argentina
Certifications
Chartered Financial Analyst (CFA) Level 1
CFA Institute
Skills
Libraries/APIs
PyTorch, ArcGIS
Tools
GIS, Sublime Text 3, Git
Languages
Python, SQL
Platforms
Jupyter Notebook, Amazon Web Services (AWS), Visual Studio Code (VS Code), AWS IoT, Google Cloud Platform (GCP)
Storage
PostgreSQL, Google Cloud
Other
Statistics, Statistical Modeling, Forecasting, Causal Inference, Machine Learning, Finance, Data Science, Generalized Linear Model (GLM), A/B Testing, Product Development, Data Warehousing, Tracking, Microsoft 365
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring