Posts

Posted on

Short course: Building a Neural Network, Code Preparing for Multi-GPU HPC and Running Large-Scale Training

University of Montenegro, a member of NCC Montenegro team, is organizing a short training dedicated to students, young researchers and professionals from industry, willing to learn about using HPC in their work, through a practical example. After learning how to create a simple neural network, training participants will be trained to prepare local environment for the development and then to copy and run the code on HPC, thus enabling model training on multi-GPU HPC.

  • Date: 12.12.2025 at 12:00h
  • Venue: Faculty of Science and Mathematics, University of Montenegro, Room 210
  • Title: Training on Building a Neural Network, Code Preparing for Multi-GPU HPC and Running Large-Scale Training
  • Designed for: students, researchers, and professionals with basic Python knowledge
Short course on Neural networks using Multi-GPU HPC and Running Large-Scale Training

Training content overview

  • Creating simple neural network for defect detection in manufacturing (1h)
  • Explaining docker containerization tool, and preparing local environment for development (2h)
  • Copying local environment to HPC (0.3h)
  • Running model training on multi-GPU HPC (1.2h)
Posted on

Modern Conversational AI – From Classic NLU to LLMs

On 21.11.2025, NCC Montenegro successfully delivered a short course as part of the EUROCC 2 and EUROCC4SEE initiatives. The program brought together an excellent cohort of students, researchers, and industry professionals who demonstrated remarkable curiosity, teamwork, and practical problem-solving skills throughout the training.

The course was delivere by mr Dejan Babic and mr Ivan Jovovic

The course explored the evolution of conversational AI, beginning with traditional natural language understanding (NLU) approaches based on intents and entities, and progressing toward modern Large Language Model (LLM) architectures and Retrieval-Augmented Generation (RAG) systems. Participants were introduced to prompt design, tool and function calling, and essential aspects of safety, privacy, and guardrails in AI systems. The curriculum also covered embeddings, vector indexes, hybrid search techniques combining BM25 with dense vectors, and re-ranking strategies for improving retrieval quality.

A significant component of the course was a hands-on laboratory session where participants built a small RAG-based chatbot using domain-specific documents. The HPC perspective was also highlighted, including batch embedding generation, large-scale indexing considerations, and methods for stress testing AI pipelines. The course concluded with live demonstrations using Azure AI Foundry, showcasing Prompt Flow, Evaluate, and AI Search capabilities.

There was around 20 participants in the event

Participants quickly absorbed the theoretical concepts, engaged with thoughtful and challenging questions, and worked independently during practical sessions. By the end of the course, they delivered functional prototype systems featuring grounded answers and clear evaluation reports—demonstrating both strong technical understanding and applied competence.

Posted on

Short course: Modern Conversational AI — From Classic NLU to LLMs

This short course covers the foundations of conversational systems—classic NLU (intents, entities, slot filling, dialogue design) and modern LLM workflows (prompt engineering, function calling, RAG). Participants build a practical chatbot grounded in their own documents, evaluate quality and safety, and deploy a lightweight interface. An HPC module is included for large-scale embeddings and offline evaluation/load testing.

  • Date: 21.11.2025 at 11:45
  • Venue: PS, UDG
  • Registration required: https://forms.gle/SRW6GYiRAbi8pFBe8
  • Designed for: students, researchers, and professionals with basic Python and web/API skills.
Short course on NLP and LLMs

Course content overview

Session 1 (90 min) – theoretical framework

  • From classic NLU (intents/entities/slots) to LLM “agents”
  • Dialogue design: state machines vs. tools/functions
  • RAG essentials: indexing, chunking, hybrid search, source citations
  • Evaluation & safety: relevance/groundedness, moderation, PII
  • HPC view: when batch embeddings and batch evaluation matter

Session 2 (90 min)- hands-on lab

  • Project setup and starter RAG pipeline
  • Document import/index, prompt + function calling
  • Quick evaluation and guardrails
  • Deploy a web chat

Learning outcomes

  • Contrast intent-based vs. LLM-based chatbots.
  • Design dialogue and implement a grounded RAG pipeline with citations.
  • Ship a lightweight production chatbot with evaluation and safety.
  • Apply HPC techniques to scale embeddings and offline performance testing.
Posted on

EuroCC Researchers Participate in the Montenegrin Machine Learning Workshop

Researchers from EuroCC Montenegro participated in the Montenegrin Machine Learning Workshop, a one-day event aimed at popularizing topics related to Machine Learning (ML) and Artificial Intelligence (AI) among students, researchers, and practitioners. The workshop was organized in cooperation with the Montenegrin Artificial Intelligence Association (MAIA) as a satellite event to the EEML summer school. Participants attended lectures covering deep learning and its applications in earth observation, graph neural networks, power grids, biology and genomics.

During the poster session, visitors had the opportunity to learn more about the EuroCC project and its efforts to promote AI and HPC research in Montenegro.

Posted on

Two AI Short Courses Successfully Completed by NCC Montenegro

Over the past two weeks we delivered two focused courses, conducted under the EUROCC 2 & EUROCC4SEE project, with an outstanding cohort of participants — students, researchers, and professionals who excelled in curiosity, teamwork, and results.

Dejan Babic giving presentation on CV &CNN supported by HPC

Computer Vision & CNNs with HPC – Short Course

  • From raw pixels to features and robust visual representations
  • Hands-on lab: building and training an image classifier
  • Running experiments on the NCC Montenegro HPC cluster
  • Participants mastered concepts quickly, asked sharp questions, and worked independently in the lab
Ivan Jovovic, giving a presentation on Edge/AI supported by HPC

EdgeAI – Artificial Intelligence & the Internet of Things supported by HPC

  • Designing efficient AIoT data pipelines
  • Deciding when to process at the edge vs. in the cloud
  • Deploying lightweight ML models on resource-constrained devices
  • Model optimization using HPC infrastructure
Demonstration by Elvis Taruh and Ivan Jovovic on running HPC created models on NVidia Jetson platform

Stay tuned for the next sessions and advanced workshops!

Posted on

HPC Development for Very-High-Resolution Atmospheric Reanalysis in Montenegro

The Institute of Hydrometeorology and Seismology of Montenegro successfully secured HPC access from the EuroHPC JU Development Call for their project titled:
“HPC Development for Very-High-Resolution Atmospheric Reanalysis Using a Nonhydrostatic Mesoscale Model over Montenegro (1995–2024)”. The project aims to enhance mesoscale weather modeling capabilities using the WRF-NMM (Nonhydrostatic Mesoscale Model) and to investigate the scalability and performance of nonhydrostatic dynamic cores on state-of-the-art high-performance computing (HPC) architectures.

Through this initiative, the project was granted 4,000 node hours on the LUMI-C partition for a period of six months. The National Competence Centre (NCC) Montenegro provided support throughout the project application process.

Posted on

Representatives of NALED and Philip Morris International visited UDG, NCC Montenegro, and CoE FoodHub

Representatives of NALED and Philip Morris International visited UDG, NCC Montenegro, and CoE FoodHub

Representatives of NALED (National Alliance for Local Economic Development), Serbia, and Philip Morris International visited the University of Donja Gorica, where they presented their leading projects — StarTech, Empower Innovation, and particularly PMInnovia — aimed at connecting science and industry, promoting the concept of open innovation, and supporting researchers and innovators in developing ideas to improve products, processes, and the competitiveness of the economy.

During the visit, the representatives of NALED and PMI were introduced to the research and project activities of UDG, the National Competence Center for Supercomputing (NCC Montenegro), and the Center of Excellence for Digitalization of Food Safety Risk Assessment and Precision Certification of Food Product Authenticity (CoE FoodHub), particularly in the areas of HPC and AI technology applications and support for digital innovations in smart agriculture, personalized medicine, monitoring the authenticity of Montenegrin food products, and bioinformatics for genomic profiling.

The delegation also toured the Virtual Reality (VR) and 3D Printing Laboratory, where student projects showcasing the connection between technology, robotics, research creativity, and industrial relevance were presented. They also visited the Food Safety and Quality Laboratory, where modern methods and equipment for physico-chemical and microbiological analyses were demonstrated, as well as the Entrepreneurial Nest, where they learned about business initiatives supporting innovation and students, such as UDG’s StartUp program and the Entrepreneurial Ideas Exchange.

The professional discussion focused on mapping regional expertise, biotechnological innovations, and exploring potential collaboration within interdisciplinary teams sharing a common vision of advancing the innovation ecosystem and the economic development of the region.

Privacy Policy Proudly powered by WordPress