Thrasyvoulos Giannakopoulos

PhD candidate in Cybersecurity & Artificial Intelligence @ UniPi

Speaker's Bio

Thrasyvoulos Giannakopoulos received his B.Sc. in Informatics and Telematics from the Harokopio University of Athens in 2019, with his thesis titled “Email Security Study”, and his M.Sc. in Digital Systems Security from the Department of Digital Systems, University of Piraeus in 2023, with his thesis titled “Threat Categorization on CVE Descriptions Using Text Classification”. He is currently pursuing his PhD in the areas of Cybersecurity and Artificial Intelligence. Since June 2021 he has been a security researcher for the Systems Security Laboratory (SSL) of the University of Piraeus, from 2025 – 2026 as a researcher for PRIVACT P.C., and since 2026 a security researcher for the National Cybersecurity Authority of Greece as a contractor. He has participated as a security analyst in the “CitySCAPE”, “RE-SAMPLE”, “NG-SOC”, “EU-iNSPIRE”, “3D-CIRCULAR”, “AKADIMOS”, and “CADMUS” European research projects within the implementation and innovation programs “Horizon 2020” and “Digital Europe” of the European Commission. His research focuses on machine learning, security, and privacy, where he has authored 5 scientific publications so far. He holds the ISO 27001:2022 and 42001:2023 Lead Auditor certifications (TÜV Austria).

Thrasyvoulos Giannakopoulos received his B.Sc. in Informatics and Telematics from the Harokopio University of Athens in 2019, with his thesis titled “Email Security Study”, and his M.Sc. in Digital Systems Security from the Department of Digital Systems, University of Piraeus in 2023, with his thesis titled “Threat Categorization on CVE Descriptions Using Text Classification”. He is currently pursuing his PhD in the areas of Cybersecurity and Artificial Intelligence. Since June 2021 he has been a security researcher for the Systems Security Laboratory (SSL) of the University of Piraeus, from 2025 – 2026 as a researcher for PRIVACT P.C., and since 2026 a security researcher for the National Cybersecurity Authority of Greece as a contractor. He has participated as a security analyst in the “CitySCAPE”, “RE-SAMPLE”, “NG-SOC”, “EU-iNSPIRE”, “3D-CIRCULAR”, “AKADIMOS”, and “CADMUS” European research projects within the implementation and innovation programs “Horizon 2020” and “Digital Europe” of the European Commission. His research focuses on machine learning, security, and privacy, where he has authored 5 scientific publications so far. He holds the ISO 27001:2022 and 42001:2023 Lead Auditor certifications (TÜV Austria).

2026
Community Presence

UniPi

Sat 21 Nov 2026 • 14:00 (EET)
Open Space
2026
Community Presence

UniPi

Fri 20 Nov 2026 • 10:00 (EET)
Open Space
2026
Community Presentation

Adversarial Attacks in Real Life: Examples from Multimedia Applications

Fri 20 Nov 2026 • 15:00 (EET) • Basic
Open Space

The security of AI systems rests on trust in their training data, inputs and components. This presentation examines adversarial threats to multimedia AI applications (that better illustrates attacks against AI) through three case studies, each with a distinct attacker objective. The first traces deepfakes from research prototypes to instruments of geopolitical deception. The second explores how concept poisoning alters text-to-image models and spreads to derivative systems. The third examines how access to model outputs or internal representations exposes training data and enables model replication. Together, the cases show how attacks across the lifecycle reinforce one another, and why assessments must state what adversaries can observe or influence.

Mentoring Slot

1:1 mentoring sessions with experienced professionals from the Open Conf community. Pick a mentor below and book your slot: https://open-conf.gr/mentors-booking/

Mentoring Slot

1:1 mentoring sessions with experienced professionals from the Open Conf community. Pick a mentor below and book your slot: https://open-conf.gr/mentors-booking/

Mentoring Slot

1:1 mentoring sessions with experienced professionals from the Open Conf community. Pick a mentor below and book your slot: https://open-conf.gr/mentors-booking/

UniWA: Blockchain Application in the Food Industry

WE LEAD Presentation

WE LEAD is an independent nonprofit organization empowering women to enter, grow, and lead in technology and innovation. During this community presentation, we will introduce our mission, programs, and impact, while sharing how we support women through free tech training, mentorship, career opportunities, and access to a strong professional network. Attendees will also discover how they can connect with and become part of the WE LEAD community.

Agenda:
• Introduction to WE LEAD and our mission
• How we support women entering, growing, and leading in technology
• Our programs, initiatives, and impact
• Ways to join and engage with the WE LEAD community

Starting a PostgreSQL User Group from Scratch

PUG Greece was launched in November 2025 with a simple goal: bring PostgreSQL users and enthusiasts in Greece together. By OpenConf, we will have organised three meetups featuring strong technical talks and built a community that is growing organically both online and offline.

This presentation tells the story of those first months—not as a polished success story, but as an honest look at what it takes to get a new PostgreSQL community moving.

An interactive workshop examines how AI reshapes professional identity, ownership, and responsibility, helping participants clarify their unique value and skills for an AI-driven workplace.

Learn how to use AI tools and plain English to turn an idea into a working website or app prototype without prior coding experience.

A hands-on workshop on using modern AI tools with WordPress for design, development, site management, and context-aware user experiences.

An interactive workshop examines how AI coding assistants and autonomous agents are reshaping software development, developer roles, and the skills needed to remain relevant.

A hands-on workshop on UI/UX for apps built with AI: build a first screen, critique it together, and fix the generic AI look.

Architectural Katas give teams repeated practice designing systems, defending decisions, and learning from alternative approaches to develop architectural judgment.