Meet the experts shaping HIVEMIND_

IDENER is a private research company specialising in the multidisciplinary field of computer science and its practical applications in optimising systems. These include chemical and biochemical systems, logistics management, manufacturing systems, energy markets, and defence. IDENER implements an extended interpretation of AI which fuses machine learning, systems modelling, advanced control and mathematical optimization. IDENER AI team on Natural Language Processing (NLP), Generative AI (GenAI), Explainable AI (XAI) and Machine Learning Operations (MLOps).

IDENER is the administrative and technical coordinator of HIVEMIND, leading the implementation of multiple specialisation methodologies for large language models (LLMs), such as fine-tuning, retrieval-augmented generation (RAG), and reinforcement learning with human feedback (RLHF). Additionally, the company is developing two agents within the HIVEMIND Framework: the CodeGenAgent and the DeployAgent.

The German Research Center for Artificial Intelligence (DFKI), founded in 1988, is a leading research institution in Germany specialising in innovative commercial software technology using Artificial Intelligence. With research facilities across multiple cities in Germany and a strong focus on application-oriented basic research, DFKI develops prototypes, product functions, and patentable solutions in information and communication technology. It conducts research through 29 departments, ten competence centres, and eight living labs, with funding from government bodies such as the EU and the German Research Foundation, as well as industrial partners. DFKI’s Speech and Language Technology department (SLT) contributes to HIVEMIND, focusing on AI for spoken and written language, information extraction, knowledge technologies, and chatbots. SLT is particularly known for its work on sparse data training, explainable AI, and multilingual resources, as well as its use of crowdsourcing and hybrid human-machine intelligence processes.

DFKI leads three of the HIVEMIND work packages. DFKI is responsible for the workflow for the collection, preparation and curation of end users data (including the development of a data gathering and data preprocessing pipeline) and also for the management of and coordination among the different use cases and use case partners involved in the project.

The University of Alicante (UA) has a strong legacy in teaching, research, and knowledge transfer. It has long been a benchmark in university extension, culture, internationalization, and territorial development. UA focuses on retaining and attracting talent, enhancing teaching quality, fostering research excellence, and promoting knowledge transfer.

On behalf of UA, the Language Processing and Information Systems Group (GPLSI) is a research group specializing in Natural Language Processing (NLP) and Human Language Technologies (HLT), with a particular emphasis on creating high-quality training resources and developing language models for a variety of NLP tasks.

Within the project, they bring expertise in human-in-the-loop approaches, with a primary role in leading Task 9.3. (Creation of HCI guidelines for human-assisted DRL for AI technologies that elicit responsible SE practices), and Task 12.4. (Workforce training and capacity building), with the aim of creating Human-computer interaction (HCI) guidelines to assist LLMs in a multi-agent environment, generating learning resources and training models. In addition, they provide support and participate in tasks related to several work packages, including WP1, WP3, WP4, WP8, WP9, WP10, WP11, WP12, WP13 and WP14.

The Universitat Politècnica de Catalunya · BarcelonaTech (UPC) is a public institution dedicated to research and higher education in engineering, architecture, sciences, and technology. As one of the leading technical universities in Europe, UPC annually graduates nearly 6,000 bachelor’s and master’s students, over 300 doctoral students, and 1,600 continuing education students. The university boasts a high graduate employment rate, with 95% of its graduates securing jobs, and 88% finding employment within three months. UPC is consistently ranked in major international rankings

UPC leads WP5 (“Smart system specification”) that will allow intelligent system specification through the development and implementation of three AI agents. In addition, UPC participates in WP1 (“Requirements, use case needs and architecture design”), WP3 (“Multi-agent collaborative architecture for responsible software engineering”) and WP8 (“Multi-agent framework integration and validation”).

RISE is Sweden’s leading research institute and innovation partner, supporting the competitiveness of the Swedish business community through international collaborations with industry, academia, and the public sector. With nearly 3,300 employees, RISE engages in and facilitates various innovation processes, offering unique expertise and over 130 testbeds and demonstration environments for future-proof technologies, products, and services. As an independent, state-owned research institute, RISE plays a key role in advancing a sustainable society.

Contributes to the development of AI agents to support design-by-contract programming during the software development lifecycle as well as accelerating the maintenance of software systems. Also contributes to the cybersecurity assessment and quality control of the solution.

An organisation built on a European model, with unique competence centers and internationally competitive specialists, SOFTWARE IMAGINATION & VISION is the Romanian software company that provides IT services directly to the European Commission organisation. The team’s experience is vast, taking over all software development activities in: education, health, customs, nuclear, business, banking, utilities, production, etc.

SIMAVI is leading WP8, which focuses on integrating and validating all technologies developed in the project to establish the complete HIVEMIND framework. This involves creating integration methodologies, ensuring cybersecurity, and refining the final framework through penetration testing and vulnerability assessments.

Additionally, SIMAVI plays a key role in WP10 by validating one of the project’s five use cases. We also have contributions in several tasks such as integrating methodologies for customizing open-source LLMs, validating a self-collaborative multi-agent architecture, and developing AI-driven agents like CodeGenAgent (for software development acceleration), MaintainAgent (for vulnerability detection and auto-repair), and DeploymentAgent (for AI-powered resource management in multi-architecture systems).

Queen’s University Belfast with over 170 years of proud history, a member of the prestigious Russell Group, is a global leader in research excellence, ranked joint 1st in the UK (Complete University Guide 2022) and 14th for research quality (Times and Sunday Times Good University Guide 2022). With over 75% of its research rated internationally excellent or world-leading, Queen’s drives innovation across disciplines, from cancer studies to computer science, future mobility (connected automated mobility systems), software engineering and cyber security. Committed to impactful, interdisciplinary research aligned with UN Sustainable Development Goals, the University fosters global collaborations to tackle local and global challenges, ensuring its research transforms lives and shapes the future.

The QUB group will be mainly involved in the use-cases work package, focusing on developing a prototype for autonomous driving using HiveMind’s AI-based approach and the Autoware framework. This will include software for integrity monitoring of perception and localisation functions, crucial for safety and trust. The team will leverage HiveMind’s AI tools to assist in software engineering tasks, from design to deployment.

HAVELSAN is a leading Turkish technology company specialising in software-intensive systems for defence, security, and civilian applications. The company focuses on dual-use technologies, developing solutions that serve both military and civilian sectors, including command and control systems, smart cities, cybersecurity, and digital transformation. In addition to its defence capabilities, HAVELSAN contributes to civilian technologies, offering public safety, healthcare, and transportation solutions that enhance digital infrastructure and operational efficiency. With expertise in big data, AI, cloud computing, and simulation technologies, the company provides advanced emergency response systems, smart grids, and digital twin solutions, driving technological progress in both defence and civilian sectors.

In the context of the HIVEMIND project, HAVELSAN builds on its experience with projects like TeamAware, which focuses on improving situational awareness through advanced sensor integration and AI-powered decision support systems. This expertise is applied to the HAVELSAN use case in HIVEMIND, which highlights disaster resilience, system resilience, and collaborative multi-agent architectures, ensuring secure and efficient software development and deployment. Through its mission-aligned contributions, HAVELSAN enhances the project’s focus on responsible, human-centric AI and software engineering practices in the disaster response and resilience perspective.

The Fraunhofer Institute for Machine Tools and Forming Technology IWU is your innovation-driven partner for research and development in production engineering. Around 670 highly qualified employees work at our locations in Chemnitz, Cottbus, Dresden, Leipzig, Wolfsburg, and Zittau. We unlock potential for competitive manufacturing, across industries like automotive, aerospace, electrical engineering, and precision engineering. Our research and contract research encompass the entire manufacturing ecosystem, from individual components to processes, methods, complex machine systems and human interaction.

Fraunhofer IWU is building complex digital twins for manufacturing process optimization serving as a use case partner for tight integration of the HIVEMIND multi-agent framework into ongoing software development efforts. Fraunhofer IWU will support the definition and validation of requirements with a focus on software development in the modern manufacturing field.

Tiga Information Technologies (TIGA), a leader in healthcare IT since 2007, is known for its patient-centric solutions and large-scale healthcare IT projects across Türkiye, Qatar, the UK, Saudi Arabia, and Estonia. TIGA’s solutions meet global standards, including ISO 27001, 9001, and 14001, and CMMI Level 3 certification. With a focus on digital healthcare transformation, TIGA has impacted over 125 million people through key projects such as the National Healthcare System (NHS) and Pharmaceutical Track & Trace System (PTTS). Their R&D efforts include AI innovations, personalised healthcare, and data analytics, with projects in population health management, mobile physiotherapy, elder care, and cancer drug preparation. TIGA’s work is supported by prestigious organisations like the European Union Programs, TÜBİTAK, and the European Commission.

TIGA serves as a use case owner, leading the healthcare-focused validation of the framework. The team supports the definition of end-user requirements, with a strong emphasis on security and data privacy. TIGA’s main role involves collaborating in the deployment and validation of the HIVEMIND multi-agent framework within a real-world healthcare scenario—specifically, a secure-by-design AI-powered software for auto-contouring organs in medical images.

AUSTRALO is a marketing agency specialising in science and deep tech, helping early-stage ideas with transformative potential thrive from Lab to Market. With a strong track record of 43 European projects since 2020, they provide services in dissemination, exploitation, and communication, working across Industry 5.0, Fast-Track innovation, and Deep Tech sectors. Their portfolio has secured over €200 million in funding and includes 500+ partners, such as top universities, startups, and Forbes 2000 enterprises. AUSTRALO’s services include promoting project visibility, facilitating innovation management, identifying strategic partnerships, and enhancing problem-solving through Design Thinking.

AUSTRALO leads WP11 and WP12 on Communication, Dissemination, and Exploitation, ensuring effective outreach, stakeholder engagement, and impact maximization. The team develops strategic guidelines, manages digital presence, and drives strategic communication for key audiences while fostering dissemination to maximize knowledge uptake. AUSTRALO also manages IPR and works with partners to develop tailored strategies for protecting and exploiting project results.

MDU is a public university recognised for its research-intensive academic approach, with strong connections to industry and regional development. With campuses in Eskilstuna and Västerås, it offers a wide range of courses and programmes across disciplines such as Design and Communication, Economics, Healthcare, Education, Engineering, and the Arts, serving around 18,000 students. MDU is dedicated to providing high-quality education and conducting impactful research, particularly in fields such as Artificial Intelligence, energy and environmental engineering, industrial technology, and health and welfare. The university offers bachelor’s, master’s, and doctoral programmes, aiming to make a difference in society through its research and academic contributions.

The role in HIVEMIND: Responsible AI in software development:

• Fairness and Bias Mitigation – AI models are designed in HIVEMIND to minimize biases, promoting fairness across diverse user groups i.e. gender and preventing discrimination in decision-making.
• Transparency and Accountability – explainability of AI models, and accountability of developers, ensuring that AI-driven decisions are understandable and traceable.
• Ethical Decision-Making and Compliance – enforce ethical guidelines, regulatory compliance, and continuous monitoring to prevent unintended consequences.

The University of Applied Sciences and Arts of Southern Switzerland (SUPSI) is one of the nine professional universities recognized by the Swiss Confederation. Within SUPSI, the Dalle Molle Institute for Artificial Intelligence (IDSIA) is involved in HIVEMIND. IDSIA is internationally recognized as a leading center in AI research, focusing on both basic and applied studies in artificial intelligence. The institute collaborates closely with industry and institutions, producing significant spin-offs that benefit both the economy and various industrial sectors.

SUPSI leads the development of the multi-agent collaborative framework for responsible software engineering, defining the foundational principles for implementing responsible AI agents by design. SUPSI develops the interaction mechanisms of AI agents, establishing the necessary interaction patterns, protocols and designing the required APIs to facilitate seamless agent collaboration. Together with validation and testing which ensures reliability and effectiveness of the multi-agent ecosystem, reaching ambitious HIVEMIND KPIs, SUPSI actively monitors the landscape of multi-agentic AI frameworks to identify novelties and emerging approaches that could enhance the system’s capabilities. The multi-agent collaborative ecosystem implemented by SUPSI serves as a backbone of HIVEMIND, allowing for multi-modal LLMs customization to build specialist AI agents, which SUPSI also supports by defining behavioral patterns of software engineering agents.