Pedro Mercado
Senior Applied Scientist at Amazon Web Services (AWS)
Germany
Summary
🤩
Rockstar🎓
Top SchoolPedro Mercado is a Senior Applied Scientist with eight years of experience combining a PhD-level background in machine learning from Saarland University with production-focused forecasting work at AWS. At AWS he has driven probabilistic time-series and forecasting capabilities for the AI/ML and Forecasting team, contributing significant model implementations and hierarchical reconciliation methods to the well-known GluonTS open-source library. His work bridges research and engineering—developing novel predictors like MovingAveragePredictor, improving robustness to missing data, and adding metric evaluation tooling that eases model adoption in production. Pedro’s trajectory includes academic research at Tübingen and Saarland on spectral methods and nonlinear eigenproblems, reflecting deep theoretical foundations applied to real-world signal and temporal problems. Based in Germany, he combines rigorous math (BSc in Applied Mathematics) with hands-on ML engineering at scale, making him adept at translating complex probabilistic models into reliable forecasting systems.
8 years of coding experience
4 years of employment as a software developer
Master of Science (M.Sc.) Computer Science, Master of Science (M.Sc.) Computer Science at International Max Planck Research School for Computer ScienceDoctor of Philosophy - PhD Machine Learning - Computer Science, Doctor of Philosophy - PhD Machine Learning - Computer Science at Universität des SaarlandesBachelor of Science (B.Sc.) Applied Mathematics, Bachelor of Science (B.Sc.) Applied Mathematics at Instituto Tecnológico Autónomo de México
English, German, Spanish
Github Skills (14)
forecasting10
mxnet10
pytorch10
machine-learning10
forecast10
time-series10
python10
numpy9
data-science9
neural-network9
pandas8
statistics8
aws7
deep-learning7
Programming languages (1)
Github contributions (5)
awslabs/gluonts
Jul 2020 - Jan 2023
Probabilistic time series modeling in Python
Role in this project:
ML Engineer & Data Scientist
ML Engineer & Data ScientistContributions:98 reviews, 9 commits, 47 PRs in 2 years 6 months
Contributions summary:Pedro primarily contributed to the implementation and enhancement of time series forecasting models within the GluonTS framework. Their work includes the development of new models like `MovingAveragePredictor`, addition of estimators to various trivial models, and the integration of hierarchical time series reconciliation methods (MinT, ERM). Furthermore, they added functionality for evaluating metrics and handling missing data in the context of time series forecasting.
forecastingpythontime-series-analysistimeseries-forecastingaws
melopeo/gluon-ts
Jul 2020 - Aug 2024
Probabilistic time series modeling in Python
Contributions:141 pushes, 59 branches in 4 years 1 month
pythontime-series-analysismachine-learningprobabilistictime-series
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.