Charilaos Mylonas
Lead AI Engineer
Biography
I’m a hands-on AI engineering lead with a Ph.D. from ETH Zurich and over a decade across machine learning, scientific computing, and software engineering. I combine research depth with ownership of the architecture, evaluation, security, and delivery of AI systems.
Some highlights of my work:
- Agentic AI at EPAM: Architected the backend framework for portfolio-focused AI agents, led legacy-system integrations, and owned hybrid-cloud authentication and authorization. Built an evaluation and benchmarking library that enabled systematic reliability testing and a 35% reduction in token consumption.
- Graph ML at Modulai: Computed graph embeddings across a client’s 35M+ users and extended its fraud-detection pipeline with scalable graph-derived features, reducing false-positive rates by up to 7% on test data.
- Research and open source at ETH Zurich: Developed graph neural networks and probabilistic generative models for monitoring engineering systems, and built and open-sourced a message-passing GNN library.
My leadership is grounded in hands-on engineering: setting technical direction, building reusable foundations, and helping others deliver. At Deloitte, I combined client-facing financial-risk and AI work with mentoring, engineering-practice improvements, and ML knowledge-sharing across North and South Europe.
Outside work, I make music and tinker with microcontrollers and software automation. You’ll find some of my personal projects below.
PhD Machine Learning for SHM under Uncertainty, 2021
ETH Zurich
MSc Computational Science & Engineering, 2015
ETH Zurich
MSc in Civil Engineering, 2012
Aristotle University of Thessaloniki
Featured Publications
Recent Posts
Table of Contents
Using knowledge graphs for drug discovery
Please refer to this repo: mylonasc/drug-target-interaction-gnns
(check out my personal blog here for other projects)
Table of Contents
Literature Clustering for Workshop
I co-organized some small workshops in my department (Civil and Environmental Engineering) called “DSML@DBAUG” where researchers (Ph.D. and Post Docs) from all chairs of the civil and geomatic engineering at ETH shortly presented their work. Since this was happening during the pandemic, everything happened through zoom. In order to be able to interact during the event, “breakout” sessions were scheduled where loosely moderated discussions around pre-defined topics were conducted among small groups.
Events
Recent Publications
Experience
Chair of Structural Mechanics and Monitoring. Research topics:
- Generative Models for UQ in engineering
- Graph Neural Networks
- Wind turbine and wind farm simulations