Skip to content
View nepiskopos's full-sized avatar

Block or report nepiskopos

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
nepiskopos/README.md

πŸ‘‹ Greetings visitor, I'm Nicholas and I am an AI (Software) Engineer!

πŸŽ“ Senior AI Engineer Β  | Β  AI Consultant Β  | Β  Agentic AI & RAG Specialist
πŸ“ Athens, Greece Β |Β  Open to Full-Time & Freelance/Contract Work

WebsiteΒ Β Β Β Β Β  LinkedInΒ Β Β Β Β Β  GitHubΒ Β Β Β Β Β  ORCID


πŸ’Ό About Me

I'm a Senior AI Engineer with 9+ years building and shipping production software systems β€” from research prototypes to enterprise-scale deployments. I specialize in Agentic AI (LangGraph), Retrieval-Augmented Generation (RAG) pipelines on Azure AI Search, and MLOps, working across Healthcare, Cybersecurity, and Banking domains.

Currently, I'm Senior ML Engineer, AI Consultant & Software Developer at Grant Thornton, leading enterprise AI transformation for Greece's Independent Authority for Public Revenue (IAPR), and Senior Software Engineering Consultant at EETAA, driving government modernization through enterprise-scale systems migration. I've also led AI transformation efforts for the National Bank of Greece as an Enterprise AI Banking Solutions Architect.

Over my career I've delivered 50+ AI projects, contributed to 10+ EU-funded research initiatives (Horizon Europe, Horizon 2020, FP7), and remain deeply passionate about applying AI and Data Science responsibly for meaningful, real-world impact.


πŸ“Š GitHub Stats

Metrics


πŸ”§ Tech Stack & Tools

  • Artificial Intelligence: RAG, AI Agents, Generative AI, LLMs, Prompt Engineering, Machine Learning, Deep Learning, NLP, Federated Learning, Neural Networks
  • Frameworks & Languages: Python, LangGraph, LangChain, TensorFlow, PyTorch, Transformers, FastAPI, scikit-learn, REST APIs
  • Data Science & Engineering: SQL, Apache Spark, PySpark, Data Analysis & Visualization, Tableau, Power BI, Data Mining
  • Cloud & Infrastructure: Microsoft Azure, Azure AI Foundry, Azure AI Search, Azure Databricks, Azure OpenAI, Azure Functions, Docker, Kubernetes, Vector Databases, DevOps

πŸš€ Experience Highlights

  • πŸ›οΈ Grant Thornton (Jan 2026 – Present) β€” Senior ML Engineer, AI Consultant & Software Developer, leading Agentic AI & ML/DL solutions for Greece's Independent Authority for Public Revenue.
  • πŸ›οΈ EETAA (2025 – Present) β€” Senior Software Engineering Consultant, orchestrating enterprise-scale legacy-to-modern systems migration for government initiatives.
  • 🏦 IBM β€” National Bank of Greece (2025) β€” Enterprise AI Banking Solutions Architect, architecting and deploying intelligent automation agents.
  • 🌍 INLECOM (2023 – 2024) β€” Delivered CONNECTOR (EU border security), EMERALDS (smart city mobility), and CRM-geothermal (critical raw materials AI) Horizon Europe projects.
  • πŸ›‘οΈ MetaMind Innovations (2022 – 2023) β€” Delivered ELECTRON, DYNABIC, TERMINET, and AI4CYBER β€” Horizon 2020/Europe cybersecurity and IoT platforms.
  • πŸŽ₯ Fogus Innovations (2019 – 2022) β€” Delivered OPTIMIST, INCOGNITO, RE-CENT, and SECONDO β€” 5G, privacy, decentralized content, and cyber-risk platforms.

🧠 Open-Source Projects

  • πŸ’¬ Mackie – Agentic AI social media assistant for nonprofit marketing teams, built on LangGraph with an agent-tools loop.
  • πŸ” LangGraph PII Detector – Advanced PII detector implemented in LangGraph, powered by Azure OpenAI.
  • πŸ“„ Document Summarization AI – Intelligent document summarization system built with LangGraph and Azure OpenAI.
  • πŸ§ͺ LLM Fine-tuning with LoRA – End-to-end guide and toolkit for fine-tuning and deploying a custom LLM assistant using LoRA/PEFT.
  • πŸ›°οΈ Federated Intrusion Detection (IEC 104) – Federated Learning-based IDS for industrial SCADA/IEC 104 network traffic, built with Flower.
  • βš™οΈ Open WebUI Enhancements – Custom tools, functions, and pipeline extensions for Open WebUI.

πŸ“š Publications

See my full publication history on ORCID.


πŸŽ“ Education

  • M.Sc. in CyberSecurity, University of West Attica
  • M.Sc. in Data Science, National & Kapodistrian University of Athens
  • University degree in Computer Science, National & Kapodistrian University of Athens

🧩 What I’m Working On

  • Agentic AI workflows with LangGraph for real-world automation
  • Enterprise RAG pipelines on Azure AI Search and Azure AI Foundry
  • Open-source LLM integrations with real-time data via RAG
  • Privacy-preserving and federated AI systems for sensitive domains

🀝 Let’s Connect

Feel free to explore my repositories, check out my portfolio site, or connect with me on LinkedIn. I’m always open to discussions on AI, research collaborations, or innovative tech initiatives.


Popular repositories Loading

  1. simple_ids_with_tff simple_ids_with_tff Public

    Training of a simple Neural Network model as an Intrusion Detection System for Cybersecurity defense using Federated Learning with the TensorFlow Federated framework.

    Jupyter Notebook 17 6

  2. simple_ids_with_flwr_iec104 simple_ids_with_flwr_iec104 Public

    Training of a simple Neural Network model as an Intrusion Detection System for Cybersecurity defense using Federated Learning with the Flower framework.

    Jupyter Notebook 16 2

  3. image-classification-textural-features image-classification-textural-features Public

    Pattern recognition image classification using textural features.

    MATLAB 2 1

  4. children_privacy_protection children_privacy_protection Public

    Forked from snikolakis/children_privacy_protection

    Machine learning pipeline for hiding small children faces in pictures to conceal their identity and protect their privacy.

    Jupyter Notebook 2

  5. dense-vs-cnn dense-vs-cnn Public

    Fully-connected vs Convolutional neural networks for multi-class image classification

    Jupyter Notebook 1

  6. yoloface yoloface Public

    Forked from sthanhng/yoloface

    Deep learning-based Face detection using the YOLOv3 algorithm (https://github.com/sthanhng/yoloface)

    Python 1