Konstantinos Nikolakakis

Machine Learning Scientist & Engineer

Welcome! I currently work as a Machine Learning Scientist at SES AI. My responsibilities include development and testing of AI models for molecular property predictions, generative AI for electrolyte synthesis, data collection and analysis, benchmarking of novel and existing AI models, and full-stack AI model deployment on cloud platforms.

Get in Touch

Research Interests

🧠

Graph Neural Networks

Learning from graph-structured data and molecular representations

🏗️

Foundation Models

Large-scale models for molecular property predictions

🔬

Generative AI

Novel materials and electrolyte synthesis using AI

Scalable MoE Methods

Efficient mixture-of-experts for big data applications

Professional Experience

Machine Learning Scientist

SES AI • Woburn, Massachusetts

Sep 2024 - Present

Leading AI/ML initiatives for battery technology and electrolyte development:

  • Multi-Task Learning & Fine-Tuning of Foundation Models
  • Graph Neural Networks for molecular property prediction
  • Generative AI for novel electrolyte synthesis
  • Full-stack deployment of AI models on cloud platforms

Postdoctoral Associate

Yale University, School of Engineering & Applied Science • New Haven, Connecticut

Jul 2021 - Aug 2024

Research in machine learning, algorithmic generalization, and robustness:

  • Developed novel generalization error bounds for SGD and variants
  • Designed robust federated learning under restricted user availability
  • Introduced SGD variants achieving 3× faster LLM training (GPT-2)
  • Established statistical guarantees for quantile multi-armed bandits
  • Designed first differential privacy algorithm for quantile A/B testing

Postdoctoral Researcher

Michigan State University, Dept. of ECE • East Lansing, Michigan

May 2021 - Jun 2021

Multi-agent decision making with applications in supply chain optimization and inventory management.

Education

Ph.D. in Electrical and Computer Engineering

Rutgers University • New Brunswick, New Jersey

Feb 2016 - Apr 2021

Thesis: Learning Tree-Structured Models from Noisy Data
GPA: 3.9/4.0

Ph.D. Candidate in Computer Science

Rutgers University • New Brunswick, New Jersey

Sep 2014 - Jan 2016

Focus: Machine Learning and Pattern Recognition
GPA: 3.8/4.0

Diploma in Electrical and Computer Engineering

University of Patras • Patras, Greece

Sep 2009 - Jun 2014

Major: Stochastic Signal Processing and Communications
GPA: 8.3/10 (Top 4% of class)
Award: 1st place Programming Competition (250 participants)

Technical Skills

Python
PyTorch
Machine Learning
Deep Learning
Graph Neural Networks
Foundation Models
LLMs
Generative AI
Multi-Task Learning
Fine-Tuning
Scikit-learn
Data Science
A/B Testing
Optimization
Federated Learning
Cloud Deployment
C/C++
MATLAB
Unix/Linux

Selected Publications

FEDSTR: Money-In AI-Out | A Decentralized Marketplace for Federated Learning and LLM Training on the NOSTR Protocol

K. Nikolakakis, G. Chantzialexiou, D. Kalogerias • Preprint

Select without Fear: Almost All Mini-Batch Schedules Generalize Optimally

K. Nikolakakis, A. Karbasi, D. Kalogerias • SIAM SIMODS

Repeated Random Sampling for Minimizing the Time-to-Accuracy of Learning

P. Okanovic, R. Waleffe, V. Mageirakos, K. Nikolakakis et al. • ICLR 2024

Federated Learning Under Restricted User Availability

P. Theodoropoulos, K. Nikolakakis, D. Kalogerias • ICASSP 2024

Beyond Lipschitz: Sharp Generalization and Excess Risk Bounds for Full-Batch Gradient Descent

K. Nikolakakis, F. Haddadpour, A. Karbasi, D. Kalogerias • ICLR 2023

Black-Box Generalization: Stability of Zeroth-Order Learning

K. Nikolakakis, F. Haddadpour, D. Kalogerias, A. Karbasi • NeurIPS 2022

Quantile Multi-Armed Bandits: Optimal Best-Arm Identification and a Differentially Private Scheme

K. Nikolakakis, D. Kalogerias, O. Sheffet, A. Sarwate • IEEE JSAIT 2021

Predictive Learning on Hidden Tree-Structured Ising Models

K. Nikolakakis, D. Kalogerias, A. Sarwate • JMLR 2021

View All Publications on Google Scholar →

Get In Touch

I'm always interested in collaborating on exciting ML and AI projects, particularly in graph neural networks, foundation models, and generative AI for material science.

📍 Based in Boston, Massachusetts