Mike Dreves
Staff Software Engineer (TLM) at Google
San Francisco Bay Area United States
Summary
🤩
Rockstar🎓
Top SchoolMike Dreves is a Staff Software Engineer based in the San Francisco Bay Area with 13 years of experience building large-scale, distributed systems in Python, Java, C/C++, and Go. At Google he serves as a TLM for flagship open-source ML tooling (TFMA and TFDV), contributing backend and evaluator improvements that help move model analysis from research into production. His work spans ML pipeline orchestration, distributed metric computation, network and security systems, and infrastructure for managing Google's technical fabric. An active contributor to TensorFlow projects, he has driven enhancements to model evaluation exports, multi-output TFLite support, and evaluator configuration to align TFX with evolving TFMA defaults. He combines rigorous academic grounding in computer science, mathematics, and business with a practical knack for making complex, safety- and security-sensitive systems auditable and reliable.
13 years of coding experience
B.Sc., Computer Science, Mathematics, Business Administration, B.Sc., Computer Science, Mathematics, Business Administration at Simon Fraser University
Github Skills (10)
machine-learning10
eval10
tensorflow10
python10
evaluation10
apache-beam9
keras9
configuration-management9
testing8
json7
Programming languages (4)
Github contributions (5)
tensorflow/model-analysis
Nov 2018 - Dec 2022
Model analysis tools for TensorFlow
Role in this project:
Back-end & ML Engineer
Back-end & ML EngineerContributions:7 releases, 12 reviews, 319 commits in 4 years 2 months
Contributions summary:Mike appears to have made several changes to the TensorFlow Model Analysis tools for TensorFlow. Their commits show involvement in enhancing the export of evaluation results, adding support for filtering extracts, and updating metrics calculations and plots. The commits also indicate work related to TFLite models, particularly focused on the use of a multi-output model.
analysis-toolsmachine-learningmodel-analysistensorboardtensorflow
tensorflow/tfx
Sep 2019 - Feb 2022
TFX is an end-to-end platform for deploying production ML pipelines
Role in this project:ML Engineer
ML EngineerContributions:1 review, 27 commits, 38 comments in 2 years 4 months
Contributions summary:Mike primarily contributed to the TFX (TensorFlow Extended) project, focusing on the Evaluator component. Their work involved restructuring configuration files, updating the evaluator to support newer versions of TFMA (TensorFlow Model Analysis), and incorporating default configurations. They also addressed issues in testing and made updates to support the updated TFMA EvalConfig defaults.
deployingend-to-endml-pipelinesmlmlops
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.