Visit to Exploring – Initiating HPC Collaboration and AI Innovation

As part of its mission to strengthen high-performance computing (HPC) capacities in Montenegro, the National Competence Center (NCC) in HPC recently visited the local IT company Exploring, a growing SME known for its innovative data-driven solutions. The meeting served as an initial consultation aimed at introducing the EUROCC project, and presenting the range of support services provided by NCC Montenegro.

NCC team visited Exploring, an SME IT developer from Podgorica

During the visit, the NCC Montenegro team provided an overview of the EUROCC initiative and the ongoing efforts to enhance digital innovation through HPC, artificial intelligence (AI), and related technologies. The Exploring team was introduced to the NCC self-assessment survey, a tool designed to help companies evaluate their readiness for adopting HPC technologies. In addition, we identified a potential use case to migrate Exploring’s development processes of ML prediction models for BI, which involve handling large datasets and developing prediction models, to the computing infrastructure managed by NCC Montenegro. This move could significantly improve computing performance and scalability for their internal R&D workflows.

The meeting also explored the possibility of:

  • Launching a proof-of-concept (PoC) analysis to evaluate the benefits of HPC in a real-world scenario tailored to Exploring’s use case;
  • Joint participation in upcoming open calls and EU-supported innovation programs;
  • Establishing a long-term collaboration between Exploring and NCC Montenegro to unlock further innovation potential.

This consultation marks an important step toward bridging the gap between academic HPC expertise and the private sector’s innovation goals. NCC Montenegro looks forward to supporting Exploring in its digital transformation journey.

We concluded this consultations with an action plan for the coming months

Conference Paper: AI and HPC Transform Matrix Learning with an Intelligent Tutoring Chatbot

The paper “Transforming Matrix Problem Solving with Intelligent Tutoring Systems” by E. Trubljanin, E. Taruh, S. Cakic, T. Popovic and L. Filipovic was presented at INFOTEH conference an is published by IEEE Xplore. Researchers from the UDG with support from HPC NCC Montenegro have developed an innovative intelligent tutoring system that leverages artificial intelligence (AI) and high-performance computing (HPC) to revolutionize how students learn matrix operations. This chatbot-based solution combines Optical Character Recognition (EasyOCR) with an advanced natural language processing model (Qwen2-Math-7B-Instruct) to interpret both text and image inputs, enabling it to perform matrix operations such as transposition, addition, and multiplication while providing clear, step-by-step explanations. Supported by the university’s HPC infrastructure, the system ensures high-speed processing and real-time feedback, achieving up to 99% accuracy in matrix recognition from high-quality images. Designed with education in mind, this AI-powered tutor enhances interactivity, understanding, and learning outcomes for students tackling complex linear algebra concepts, and sets the stage for future enhancements like handwritten input recognition and support for more advanced operations.

ABSTRACT – This paper presents the integration of optical character recognition (OCR) and advanced natural language processing (NLP) models for automated handling of matrices derived from images and textual inputs, all combined within an implemented chatbot. The motivation for choosing this topic arises from the practical experiences of the authors gained while working with groups of students who encounter the concept of matrices as part of their academic responsibilities. Through the analysis of their results and classroom interactions, it was observed that many students struggle with this area. This paper presents an innovative approach to enhancing matrix problem-solving by leveraging intelligent tutoring systems supported by High-Performance Computing, aiming to improve learning efficiency and student outcomes. By combining the EasyOCR framework and the Qwen2-Math-7B-Instruct model, operations such as transposition, addition, and multiplication of matrices are enabled. The system supports the input of one or two matrices, allowing the selection of operations through textual or image-based queries. The OCR component extracts numerical data from images, while the NLP model interprets user requests and executes operations accurately. The interface allows the addition of a second matrix image only when necessary, enhancing the system’s intuitiveness and efficiency. The results of the recognition accuracy of the OCR model of image input matrices of different dimensions show a high level of accuracy of 95%, while for 2×2 matrices they reach an accuracy of 99%. This work contributes to the development of AI-powered tools for mathematical operations and holds potential applications in education.

Click on image to open

Upcoming NVidia Bootcamps in EuroCC2 & EuroCC4SEE

Want to learn how to use artificial intelligence in scientific research? Or how to write efficient parallel code for computations on multiple graphical accelerators?

Join one of the upcoming NVIDIA bootcamps – fully online, free of charge, and led by experts from the field. Registration capacity is limited – don’t wait too long!

AI for Science Bootcamp (27-28 May 2025)

  • Learn how to use tools like PyTorch, NVIDIA Modulus, and NeMo to tackle scientific problems using artificial intelligence.
  • Link: https://lnkd.in/e7h7n-52
Training on AI for Science (source EuroCC2 & EuroCC4SEE)

Multi-GPU Programming Bootcamp (17-18 June 2025)

  • Interested in performance scaling and parallel programming across multiple graphical accelerators? This course will walk you through writing efficient multi-GPU code using libraries such as NCCL and NVSHMEM.
  • Link: https://lnkd.in/ekPSGCRg
Training on Multi-GPU programming (source EuroCC2 & EuroCC4SEE)

Fain Tech Startup Launches Advanced Research Using HPC Resources

Fain Tech, a startup specializing in computer vision, with the support of the NCC Montenegro team, which is implementing the EuroCC2/EuroCC4SEE project, has successfully gained access to Leonardo BOOSTER HPC resources for a period of three months through the EuroHPC Benchmark public call. This access will support research on the Enfield: European Lighthouse to Manifest Trustworthy and Green AI project, aimed at developing advanced models for defect detection on PCBs.

Another SME to utilise European HPC resources

With these resources, the team will focus on optimizing neural networks to create solutions that perform effectively even with limited data, minimizing the risk of overfitting. The team will demonstrate their capability to effectively use advanced computing resources, contributing to the development of reliable and sustainable technologies in industrial processes.

EuroHPC JU: Big announcement for Europe’s AI Ecosystem!

Yesterday, the European Commission presented an AI Continent Action Plan to make Europe a global leader in Artificial Intelligence! As part of this, the EuroHPC Joint Undertaking (EuroHPC JU) is helping to build a large-scale AI data and computing infrastructure (source EuroHPC JU).

EuroHPC AI Factories Ecosystem (image: EuroHPC JU)

The AI Factories Portal is LIVE!
The brand new AI Factories Portal is now open! Explore each selected AI Factory, see what services they offer, and discover how they will help startups, researchers, and innovators across Europe [link].

New Call for AI Factory Antennas
The new call aims to expand the AI Factories network. These local hubs will extend the reach of AI Factories, providing support, services, and access to EuroHPC systems across Europe. The AI Factory Antennas will closely cooperate with selected AI Factories to strengthen the European AI ecosystem [link].

Public Consultation on AI GigaFactories
Stakeholders are invited to help shape the future of hashtag#AIinEurope by expressing interest in the next-generation AI GigaFactories!
These will enable training and deployment of massive AI models using next-level computing & data infrastructure [link].

Updated EuroHPC JU Access Policy
The EuroHPC JU access policy has just been updated to include new access modes!
Researchers, startups, and public actors are now able to select several types of access modes to all operational EuroHPC supercomputers and hashtag#AIFactories. These steps are part of the EU’s mission to ensure security by enabling trustworthy Al systems.

FiveG gained access to HPC resources through the support of NCC Montenegro

Five Group company has established itself as one of the leading IT companies in Montenegro in the field of 3D design, gaming industry, application of AI/ML in various business areas, for innovative IT solutions in the field of education, military industry, etc. With the support of the NCC team of Montenegro, Five Group company submitted an application for access to HPC resources on the Leonardo supercomputer, within the Benchmark call. As a result of the successful application for the project called “Wasco AI – Custom AI Assistant Optimization via HPC”, access to the Leonard Booster partition for three months, in the amount of 3500 node-hours, was granted.

Successful submission for Benchmark call “Wasco AI”

Wasco AI is an intelligent, customized AI assistant designed for business applications. It supports multimodal interaction (text, video, images, and documents) and provides adaptive support to various industries. Through this project, benchmarking, training optimizations and fine-tuning of large language models (LLM) will be performed to ensure scalable performance, using EuroHPC resources.

Join Our 2-Day Course: Web Programming with Python, Flask and HPC-enabled ML

We are excited to invite students, researchers, and professionals to our 2-day hands-on short course on Web Programming with Python, Flask and HPC-enabled Machine Learning.

Registration open until 16 Apr 2025.

This dynamic course is designed to introduce participants to web development using Python and Flask, while exploring how machine learning applications can be accelerated with High Performance Computing (HPC). Whether you’re a beginner or looking to extend your skills to include scalable ML solutions, this course is the perfect opportunity to learn from experienced instructors and work on real-life examples. More information abou t the course can be found at [link].

Course Schedule
Part 1: Thursday, April 17, 2025
Starts at 8:00 AM
University of Donja Gorica (UDG)

Part 2: Saturday, April 26, 2025
Starts at 9:00 AM
University of Donja Gorica (UDG)

What You’ll Learn

  • Fundamentals of web development with Python and Flask
  • How to design and deploy simple web applications
  • Integration of machine learning models into web apps
  • Basics of HPC and how to run ML workloads efficiently
  • Real-world applications and interactive coding sessions

Registration

  • Registration is open until April 16, 2025.
  • Secure your spot now via the following [link].