Biography

Building trustworthy, privacy-preserving, and interpretable AI for real-world deployment.

Ph.D. in Artificial Intelligence (UPC-BarcelonaTech). I work on federated learning, explainable AI, and applied deep learning — from research to containerized production systems.

Biography

Dr. Arijit Nandi is an Advanced AI researcher and engineer with a Ph.D. in Artificial Intelligence from UPC-BarcelonaTech and over 6+ years of experience building scalable, privacy-preserving, and interpretable AI systems for real-world deployment.

His work spans federated learning for privacy-preserving model development, explainable AI (XAI) for transparency and accountability, and applied deep learning across time-series, sensor data, and language-based systems. On the applied side, he builds LLM-powered tools for intelligent automation — including content personalization, workflow optimization, and decision-support assistants — alongside an XAI-as-a-Service platform aimed at democratizing model interpretability. He architected EDFL, an open data space federated learning framework recognized by the European Commission on its Innovation Radar.

He bridges academic rigor and production deployment, with experience across the full ML lifecycle: from research and prototyping to containerized, scalable systems in real-world environments. His focus is on making AI not just performant — but transparent, safe, and human-centric.

Core expertise

Machine Learning Deep Learning Federated Learning Explainable AI Large Language Models Trustworthy AI Emotion Recognition Research & Innovation
Arijit Nandi — AI research and development
Visitor map — countries that have viewed this site

Technical Skills

Core stack for modeling, data systems, MLOps, and applied AI delivery

Programming

  • Python 5+
  • MATLAB 6+
  • JavaScript 3+
  • SQL 4+

Machine Learning & AI

  • TensorFlow 5+
  • PyTorch 4+
  • Scikit-learn 5+
  • Keras 4+
  • LangChain 0.5+

Data Science

  • Pandas 5+
  • NumPy 5+
  • Matplotlib 4+
  • Seaborn 4+
  • Plotly 3+

Big Data & Cloud

  • Apache Spark 0.3+
  • Hadoop 1+
  • AWS 0.5+
  • Azure 0.5+
  • Kubernetes 1+

MLOps & DevOps

  • Docker 3+
  • Git 6+
  • MLflow 2+
  • Kubeflow 0.3+

Databases

  • PostgreSQL 2+
  • MongoDB 2+
  • Redis 0.3+
  • Kafka 0.3+
  • Airflow 0.3+

Web & APIs

  • FastAPI 2+
  • Flask 2+
  • Django 0.2+
  • React 0.2+
  • Streamlit 3+

Specialized AI/ML

  • Computer Vision 2+
  • Federated Learning 4+
  • Explainable AI 3+
  • Reinforcement Learning 0.3+

Work Experience

My professional journey

Advanced Machine Learning Researcher

Big Data and Data Science Unit Eurecat, Barcelona, Spain

Leading AI initiatives and developing machine learning solutions for enterprise clients.

  • Worked on the "AI4Drought" project in collaboration with European Space Agency (ESA), Lobelia Earth, and Barcelona Super Computer (BSC) to improve seasonal climate predictions of drought through Earth Observation data and climate variables analysis
  • Contributed to explainable AI (XAI) methods for explaining drought prediction AI models and identifying the most influencing indicators for drought prediction
  • Developed an explainable AI GUI app (XaiSS) that combines different XAI open-source libraries into a unified drag-and-drop platform for AI models (Scikit-learn, TensorFlow, or PyTorch)
  • Deployed XaiSS application using Python 3.10, Streamlit, and Docker-container for easy accessibility and scalability

Artificial Intelligence Consultant (Freelance)

Tuttify.io

Contributed as an AI expert to design and develop emotion recognition from facial expression using Tensorflow 2.0 and FER 2013, CK++. The trained model is deployed in the backend of Tuttify (https://tuttify.io/) to recognize different emotions of the students.

  • Developed emotion recognition system using TensorFlow 2.0 and FER 2013, CK++ datasets
  • Deployed the trained model in Tuttify backend for real-time student emotion analysis
  • Implemented pose estimation of students to analyze attention span and engagement
  • Combined emotional status with pose data to provide comprehensive engagement metrics

Machine Learning Researcher

Training Unit Eurecat, Barcelona, Spain

Contributed as an AI expert to design and develop AI MOOC course for young people (AIM4YOU) under the EU project YNSPEED (Youth new personal & employable skills development).

  • Designed and developed comprehensive AI MOOC course (AIM4YOU) for young people under EU project YNSPEED
  • Course available at: https://irea.teachable.com/p/artificial-intelligence
  • Developed initial version of collaborative and content-based filtering recommendation system for Moodle LMS
  • Implemented course recommendation algorithms to enhance student learning experience and engagement

Featured Projects

Research systems with published outcomes and open-source implementations

DFL - Docker-Based Federated Learning Framework
Computing (Springer, 2023)

DFL: Docker-Based Federated Learning Framework

Designed and deployed a Docker-enabled federated learning framework (DFL) for multi-modal data stream classification. Clients and global servers communicate via lightweight MQTT protocol, enabling real-time emotion state classification from distributed physiological data (EDA + RB) while preserving data privacy.

Python TensorFlow Keras Docker MQTT Federated Learning
View paper GitHub
RPWE - Reward Penalty Weighted Ensemble for Multimodal Data Stream Classification
International Journal of Neural Systems (2022)

RPWE: Reward-Penalty Weighted Ensemble

Developed a novel Reward-Penalty Based Weighted Ensemble (RPWE) approach for emotion state classification from multi-modal physiological data streams (DEAP & AMIGOS datasets). Classifiers are dynamically rewarded or penalized based on predictive performance, with an auto-adjusting beta factor.

Python Scikit-Learn Ensemble Learning Jupyter Notebook Data Streaming
View paper GitHub
Fed-ReMECS - Federated Learning for Real-time Emotion State Classification
Methods (Elsevier, 2022)

Fed-ReMECS: Federated Real-time Emotion Classification

Built a federated learning framework (Fed-ReMECS) for real-time emotion state classification from multi-modal physiological data streams via wearable sensors. Uses MQTT for IoT communication and builds a global classifier without accessing users' local data, ensuring privacy. Evaluated on the DEAP dataset.

Python TensorFlow Keras MQTT Federated Learning IoT
View paper GitHub

Education

My academic background

2019-2022

Ph.D. in Artificial Intelligence

Polytechnic University of Catalonia (UPC-BarcelonaTech)

Specialized in Artificial Intelligence and Machine Learning with focus on deep learning applications.

2017-2019

Master of Technology in Computer Science

National Institute of Technology, Durgapur, West Bengal, India

Specialized in Artificial Intelligence and Machine Learning with focus on deep learning applications.

2012-2016

B.Tech in Computer Science and Engineering

Budge Budge Institute of Technology

Graduated with honors. Focused on software engineering and data structures.

Thesis

Doctoral research

Ph.D. Thesis

Multimodal data stream classification and prediction of e-learner’s emotional states

Director
Prof. (Dr.) Fatos Xhafa · Dra. Laia Subirats Mate
Defense date
2024
Document type
Doctoral Thesis
Collection
Universitat Politècnica de Catalunya
Portal de la Recerca de Catalunya Dialnet Scilit

Publications

Research contributions and scientific work

Showing 36 publications

2025

2024

2023

2022

2021

2020

2019

Blog

Recent essays on machine learning and applied AI

09 May 2026 · Federated Learning

Federated Learning Explained Simply (With TensorFlow Code)

A friendly introduction to federated learning—what it is, why it matters for privacy, and how to simulate it with plain TensorFlow and Ke...

22 Jan 2024 · Deep Learning

Deep Learning Fundamentals: Neural Networks Explained

Explore the fundamentals of deep learning and neural networks, from basic perceptrons to complex architectures, with hands-on examples an...

15 Jan 2024 · Machine Learning

Introduction to Machine Learning: A Comprehensive Guide

This comprehensive guide covers the fundamental concepts of machine learning, from basic algorithms to advanced techniques, with practica...

All posts

Research Profiles

Scholarly indexes, identifiers, and professional profiles

Scholarly indexes

Google Scholar Publications, citations, and h-index DBLP Computer science bibliography Scopus Elsevier author profile Web of Science Clarivate researcher record ResearchGate Research network and preprints

Identifiers

ORCID 0000-0003-4238-5183 SciProfiles MDPI researcher profile

Code & professional

GitHub Open-source projects and code LinkedIn Professional background and network

Get in Touch

Research partnerships, consulting, and applied ML collaborations

Email arijit4ai@gmail.com LinkedIn Professional messages welcome
Location Barcelona, Spain
Collaboration

Interested in trustworthy AI, federated learning, or applied ML systems?

Send a brief note and I’ll reply with the best next step — whether that’s a call, a research discussion, or a scoped engagement.

  • The problem or research question you’re working on
  • Timeline, constraints, and expected outcomes
  • The kind of support you need (research, engineering, review, or advisory)
Email Arijit GitHub