Konstantinos Karanasos
Software Engineer at The Apache Software Foundation
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolKonstantinos Karanasos is a software engineer with nine years of professional experience specializing in large-scale distributed systems, query optimization, and ML systems infrastructure. He has held research and engineering roles across Microsoft (including leading the Bay Area Gray Systems Lab), IBM Research, and Meta’s Data Infrastructure org, and is an active Apache Hadoop committer and PMC member. His work spans from dynamic query optimization and cluster resource management to practical ML runtime improvements—contributing to high-profile open-source projects like ONNX Runtime by improving graph transformations and operator elimination for faster inference. Based in Mountain View, he blends academic rigor (PhD work in data management) with production-facing engineering leadership, often bridging applied research and scalable engineering. Notably, he has repeatedly moved innovations from research prototypes into cloud-scale systems, demonstrating a knack for turning complex theory into robust, deployable solutions.
9 years of coding experience
9 years of employment as a software developer
PhD student Computer Science, PhD student Computer Science at Paris-Sud University (Paris XI)Diploma Electrical and Computer Engineering, Diploma Electrical and Computer Engineering at National Technical University of Athens
Github Skills (8)
machine-learning10
rewrite-rules10
c-language10
onnx10
cprogramming-language10
deep-learning9
neural-network8
hardware-acceleration8
Programming languages (2)
Github contributions (5)
microsoft/onnxruntime
Jan 2019 - Oct 2019
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Role in this project:
Back-end Developer
Back-end DeveloperContributions:17 commits, 31 PRs, 132 pushes in 8 months
Contributions summary:Konstantinos's primary contribution focused on enhancing the ONNX Runtime's graph transformation capabilities. This involved implementing and refactoring rewrite rules, specifically targeting the elimination of Slice and Identity operators, and constant folding. The changes included updating graph utility functions and the rule-based graph transformer, indicating a focus on improving the efficiency and optimization of model execution. The user demonstrated proficiency in modifying core graph transformation components.
runtimetrainingtensorflowai-frameworkaccelerator
kkaranasos/presto
Nov 2022 - Jan 2023
The official home of the Presto distributed SQL query engine for big data
Contributions:19 pushes, 1 branch in 3 months
jdbchbasebigdatabig-datatrino
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.