Ioannis Chrysakis in Spotlight for STELAR: The Constant Evolution of AI
STELAR Project’s Data Stories 360° podcast is working on bringing you stories from the field and insights into how data and AI are being used today. The host for this episode is Sandra Kolaric, Dissemination and Communication Manager at Foodscale Hub, one of the partners in the STELAR Project.
Today, the guest is Ioannis Chrysakis, a Research and Innovation Project Manager at Netcompany-Intrasoft and a member of the GREEN.DAT.AI project, which is a sister project to STELAR. With extensive experience in research and innovation across more than 20 National and European-funded projects, his expertise spans AI, Data Privacy, Data Spaces, Digital Agriculture, and more.
In this episode, we explore the evolution of AI and data-driven solutions in agriculture, and discuss key challenges, innovations, and the future of AI in both agriculture and beyond.
Exploring Our Guest's Motivation and Background
With several years of experience in ICT project management, what initially sparked your interest in this field? Additionally, how has your career evolved to include areas such as AI and data spaces?
Working in the tech domain is surely challenging, as technology evolves at a very unpredictable pace. This makes ICT projects increasingly complex. Throughout my career, I have learned that staying innovative requires continuously following technological advancements and integrating state-of-the-art developments from different fields, such as AI.
This, in turn, makes project management even more demanding, because it involves bringing together experts from different backgrounds and disciplines. At the end of the day, we want all these experts to collaborate effectively. So yes, it is very challenging to deal with this kind of project, but I am really happy to have been involved in such innovative projects over time.
How have you seen the role of AI evolve in R&D projects? And how do you stay motivated in such a fast-evolving field like AI?
AI has evolved significantly over the past few years in R&D. We have moved from traditional machine learning models focused on automation and optimisation to advanced generative AI. This enables creativity, deeper insights, and more adaptive decision-making. AI now plays a crucial role in data-driven innovation, enhancing everything from predictive analytics to intelligent system design.
The constant evolution of AI, as described earlier, keeps me motivated. The opportunity to collaborate with experts across different disciplines, and the challenge of integrating cutting-edge AI techniques and algorithms into real-world applications, is a key driver for me. AI is transforming our lives, impacting both the personal and business spheres.
AI and Agriculture
In your opinion, what role do data-driven solutions play in building resilience in agriculture and ensuring sustainable farming?
I will give a little taste of the GREEN.DAT.AI project to my answer. Digitisation is rapidly transforming agriculture with technologies like IoT, big data analytics, and AI in order to enhance decision-making and promote sustainable farming.
In particular, in AI, we can have several key advancements. I will briefly mention three:
- predicting soil health and optimising harvesting times;
- detecting pests and diseases early through real-time object assessment coming from satellite or drone images;
- generating real-time fertilisation maps tailored to soil and crop needs for optimised fertilisation.
These are exactly the kind of solutions we are developing in the GREEN.DAT.AI project, and we directly support sustainable farming practices through our project.
Project Involvement and Contribution
Netcompany-Intrasoft is a partner in the GREEN.DAT.AI project. Could you elaborate on the company's role in this initiative and how do you contribute to the project’s goals?
The Netcompany-Intrasoft, through the Research and Innovation Development team, plays a key role in the GREEN.DAT.AI project project because it leads the integration activities and ensures efficient software development through state-of-the-art CI/CD (Continuous Integration/Continuous Development) practices.
Moreover, our contributions include the development of a workflow management engine, which is a tool designed to manage and execute complex AI workflows. We have also been adapting to the StreamHandle platform, which facilitates seamless communication and metadata exchange across the GREEN.DAT.AI components.
All these efforts are essential in enabling interoperability and enhancing the overall impact of the project in the agriculture domain and other domains as well.
What are the key challenges you face in managing large-scale AI-driven projects like GREEN.DAT.AI, particularly in terms of project management?
First of all, managing large-scale AI-driven projects like GREEN.DAT.AI comes with several challenges by default. The first thing I can briefly mention is that coordinating diverse development teams from multiple partners requires strong collaboration and alignment.
Secondly, setting up infrastructure to support various technologies and user workflows is also very complex. Beyond delivering concrete results, we also integrate innovative AI technologies that are not plug-and-play, so we need to invest extra time and a lot of effort to ensure that we have a fully functional solution.
This demands an agile approach to both development and service delivery, ensuring that the solution will efficiently cover the needs of all GREEN.DAT.AI users and the GREEN.DAT.AI AI pilot.
What specific strategies or innovations are you employing in GREEN.DAT.AI to overcome the challenges of interoperability and scalability in AI-driven data spaces?
We tackle interoperability and scalability challenges by, first of all, introducing a reference architecture that provides a robust core infrastructure for deploying generic-purpose data spaces. I had the privilege of presenting this architecture last week in an online workshop organised by our project. This architecture supports both data providers and AI service participants, ensuring seamless integration and efficient data exchange.
To achieve true interoperability, we also leverage well-established data space technologies coming from specifications like IDSA and GIA-X. These technologies create a standardised foundation for AI-driven data spaces, facilitating secure and scalable interaction.
I would also like to highlight a standout innovation in the GREEN.DAT.AI project, which is the energy efficiency framework. This framework assembles the energy footprint of machine learning algorithms. It helps us evaluate the scalability and sustainability of AI services, ensuring that the AI-powered solutions we develop in the context of the project are not only effective but also energy-efficient, which is very important as well.
Both GREEN.DAT.AI and STELAR focus on data-driven solutions, yet each has a unique angle. Can you explain how your project complements the work being done in STELAR?
The GREEN.DAT.AI project complements STELAR by integrating AI services and structured workflows to enhance data-driven decision-making. STELAR contributes value-added data-driven services, such as entity extraction, correlation discovery, and bias detection and mitigation. All these processes support, in some way, the initial stages of data transformation, which form an integral part of each identified workflow. These workflows are designed to achieve specific objectives, addressing the needs of each user group.
Of course, these workflows can be extended or executed using our GREEN.DAT.AI data workflow management engine. As a bottom line, we can say that there is a lot of common development and business, and we plan to continue our collaboration between these two projects.
Future Outlook
What are your thoughts on the future of AI and data management in agriculture?
The future of AI and data management in agriculture is incredibly promising, this is for sure. As technology advances, we see more intelligent, data-driven solutions that ultimately optimise farming practices. AI will enable real-time monitoring, predictive analytics, and more intelligent decision-making, which will help farmers enhance productivity while reducing environmental impact.
In parallel, interoperable data spaces will play a key role in securely sharing agricultural data, fostering further collaboration and transparency across the value chain. Additionally, AI-driven insights will support precision farming, climate adaptation, and resource efficiency, contributing to more resilient and sustainable agriculture.
The combination of AI and data management will empower farmers with smarter, more intelligent tools, helping them improve food security and align with sustainability goals like the Green Deal. All these technologies will benefit both the farmers and the end users.
How do you see the future of AI and data interoperability shaping broader industries beyond agriculture?
AI and interoperability are already transforming industries beyond agriculture. For example, we see that AI has already been adopted in mobility, energy, healthcare, and other domains. However, I would like to highlight that a key factor in this evolution is the adoption of common standards, which facilitate interoperability within the domain, ensuring seamless data exchange and system integration.
It is crucial to have a well-established framework coming from these standards. Equally important is the development of robust regulatory frameworks, an area where Europe is taking a very proactive approach. In my view, clear legislation is essential to ensure that responsible and ethical AI across all sectors will foster more trust, transparency, and long-term sustainability.
As AI and data interoperability continue to advance, these tools will drive and push for more innovation, efficiency, and smarter decision-making across several industries beyond agriculture.
Conclusion
In this episode, our listeners had the chance to hear from Ioannis Chrysakis about how AI and data-driven solutions are enhancing agriculture and other sectors. He shared insights on the challenges of managing AI projects, the importance of interoperability, and the role of sustainability in the development of these technologies.
Subscribe to our YouTube channel and visit our Blog for more insights on data innovation, the practical use of AI, and the ever-changing world of data sharing!
