Miguel Trejo is a Technical Account Manager at Google with 7+ years in IT networking and a long track record designing and operationalizing Data Center and Cloud strategies for large enterprises. He blends hands-on networking expertise (ACI, Nexus, UCS) and virtualization with customer-facing roles—presales, solutions architecture and technical account management—helping teams evolve services and seize upsell opportunities. Miguel pairs mentoring and team-building skills with a strong engineering background from Cisco and Public Sector engagements, driving consistency and operational optimization across complex deployments. Beyond networking, he contributes to open-source ML infrastructure—enhancing PyCaret’s time-series module and improving Feast’s DynamoDB online store—demonstrating a bridge between infrastructure and ML product needs. Known for pragmatic automation and reproducibility (MLFlow integration, batch deduplication), he surfaces technical improvements that reduce operational toil while enabling innovation. Based in Greater Mexico City, he combines telecom roots (Tecnológico de Monterrey) with a rare mix of network architecture and ML/feature-store engineering.
6 years of coding experience
18 years of employment as a software developer
Bachelor Electronics and Telecommunication, Bachelor Electronics and Telecommunication at Tecnológico de Monterrey
An open-source, low-code machine learning library in Python
Role in this project:
Data Scientist
Contributions:85 reviews, 128 commits, 18 PRs in 11 months
Contributions summary:Miguel primarily focused on adding and modifying the base timeseries module, introducing key functionalities for time series analysis within the PyCaret library. Their contributions included the implementation of setup function, designed for preparing data, and the incorporation of new features such as outlier removal and target transformation. The user integrated MLFlow tracking to monitor and log experiments, including data, plots, and data profiles for enhanced experiment tracking and reproducibility. Additional contributions included adding auto_select, save_model and load_model functions.
The Open Source Feature Store for Machine Learning
Role in this project:
Back-end Developer
Contributions:25 reviews, 7 commits, 7 PRs in 1 month
Contributions summary:Miguel primarily contributed to the implementation and enhancement of the DynamoDB online store within the Feast feature store. Their work included adding support for batch reads, introducing a configurable batch size, and implementing a mechanism to deduplicate batch write requests. The user also focused on improving the DynamoDB integration by adding support for endpoint URLs and making other general improvements to the DynamoDB functionality, including fixing bugs and adding unit tests.
pythondata-qualitydata-sciencemlmachine-learning
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.