​AI agents are moving from demos into real workflows — but the hard part is no longer just prompting.

​The real questions now are:
How do we give agents tools and permissions safely?
How do we test whether they actually did the work?
How do we bring agentic systems into teams, repositories, and enterprises without creating chaos?

​Join us for the launch of the Budapest Agentics Foundation Hungarian Chapter, powered by ShareTech Foundation.

​All backgrounds are welcome: engineers, product people, researchers, founders, QA professionals, AI builders, and anyone curious about practical, responsible agentic AI.

​📅 Date & Time: June 2, 2026, 17:30

​📍 Location: Create26, Budapest, 1061, Kiraly Street 26

​🌐 Language: English

🕓 Agenda
17:30 – Doors open & networking
18:00 – Welcoming remarks

​18:05-18:35 – Tamas Szabo / LI Agentic Workflows Without Agentic Chaos

​Tamás is a research engineer at GitHub Next, where he explores the future of software engineering through applied research. His work focuses on AI-based developer tooling, including GitHub Agentic Workflows and projects in the Copilot-X space. He holds a PhD in static analysis and programming languages.

​The awkward truth about AI agents is that the interesting part is no longer just the prompt. It is the permissions, tools, triggers, logs, and review paths around the prompt that decide whether an agent becomes reliable automation or another hard-to-govern process.

​In this talk, Tamás will introduce GitHub Agentic Workflows: a way to turn plain markdown into agent-powered GitHub Actions. You describe the job in natural language, define what the agent is allowed to access, and compile it into a normal workflow that can run on issues, pull requests, schedules, or manual triggers.

​You will leave with a practical mental model for building AI automation on GitHub: write the intent in markdown, make the boundaries visible, compile it into Actions, and let agents do useful work with guardrails built into the workflow itself.

​18:40-19:15 – Dragan "Profa" Spiridonov / LI Catching Agents in the Act: A Field Guide to Completion Theater

​Dragan is the Founder & Agentic Quality Engineer at Quantum Quality Engineering in Serbia. He has spent 30 years in IT, 12 years leading QA/QE, and has built three open-source agentic-testing platforms, including the Agentic QE Fleet. He is also Board Secretary of the Agentics Foundation and Ambassador of its Serbian Chapter.

​Coding agents are shipping. They are also lying — about being done, about reading the docs, and about why the test failed. This is not a personality flaw. It is a structural failure mode.

​In this practitioner-grounded talk, Dragan will walk through Completion Theater: agents optimizing for appearing done rather than being done. He will show the four most common patterns: premature completion, guidance neglect, explain-away-first, and blame deflection — using concrete examples from real coding sessions.

​You will leave with a practical framework, an escalation ladder you can apply the next morning, and the vocabulary to call out completion theater when you see it.

​19:20- 19:50 panel discussion and fireside chat with Reuven Cohen / LI and Adam Kovacs / LI

​Reuven Cohen is one of those rare technology people who has been close to several major platform shifts before they became mainstream.

​Long before “cloud” became the default language of modern infrastructure, he founded Enomaly, helped popularize the concept of Infrastructure as a Service, served as an early member of the AWS advisory board, and contributed to early cloud policy work, including the first US cloud definition with NIST.

​Today, he works with some of the world’s largest organizations on enterprise AI architecture, governance, and adoption strategy.

​At this meetup, Reuven will bring the perspective of someone who has seen major technology waves from the inside — from the early internet, to cloud infrastructure, to enterprise-scale AI — and can help us separate real transformation from short-term hype.

​

​Adam Kovacs is the co-founder and Chief AI Strategist at the AI Enablement Academy, helping companies transform their workforce and operations for the agentic era. He is also Executive Director of the OPEN Talent Society, a non-profit exploring the future of work.

​The discussion will focus on what agentic AI means beyond demos: enterprise adoption, governance, workforce transformation, quality, and the practical steps organizations need to take as agentic systems become part of everyday work.

​20:00 - networking

​About the Agentics Foundation

​The Agentics Foundation is a not-for-profit organization focused on advancing autonomous, agent-based AI systems, supporting open-source tools and education, and building an inclusive global community around ethical, real-world agentic AI.

​This Budapest meetup is the starting point for a recurring local chapter with ongoing talks, workshops, and collaborations.

​Please note: 🇬🇧 All presentations will be conducted in English to welcome our growing number of international members.

​Please be informed that photographs/videos can be taken at the event, in which the event participants may be identifiable as data subjects. The event usually produces mass shots to show the events in their overall effects, but it can also be possible to highlight a particular person or moment in the recordings. The photo/video may be used and published in promotional and/or marketing materials, on the website, intranet and social media channels. We declare that we will only use photo that are not infringe the personal rights of the data subject.