Budapest | Claude Code for Security
Join us for the next Claude Community Budapest Meetup, hosted by Tamas, Attila, Balazs, Balu, and Jamie.
Whether you’re already integrating Claude Code, Cowork into your daily workflow, curious about agentic coding, or simply looking to connect with others exploring AI-powered development, this event is for you.
Who Should Attend?
This meetup is open to everyone, regardless of experience level—just be curious:
Founders & Operators: Learn how to scale efficiently and move faster.
Business Professionals: Discover how to delegate repetitive tasks to AI agents.
Engineers & PMs: See firsthand how Claude can significantly reduce development cycles.
The AI-Curious: Transition from theory to practice with hands-on insights rather than just another lecture.
Agenda
05:30 PM: Doors open. Registration, Networking & Drinks
06:00 PM: Welcome Remarks
Jose Roca - Agentic Builder & Strategist | Building AI that does the job, not just the demo from 49x
Jozsef Ottucsak - Product Security Architect | AI/ML Security Expert | CISSP, MBA, ISO 27001, ISO 42001 from Diligent
Jozsef Ottucsak is a Product Security Architect with over a decade of experience in secure-by-design engineering and scalable AppSec infrastructure. Holding OSCP and CISSP certifications, his technical focus centers on hardening cloud-native agentic applications and securing autonomous LLM pipelines. His current engineering work involves designing runtime security guardrails, automated vulnerability detection, and threat modeling frameworks tailored for agent-driven software architectures.
-----------------------------------------------------------------
When dev agents code at high velocity, manual security checks just don't scale. We need to treat security as a standard non-functional requirement that sits directly inside the agent's loop.
We will chat through how to wire up security tools within agentic workflows, focusing on:
Early-stage checks: Running architecture and design reviews before the agent even writes code.
Orchestrating traditional tools: How an AI agent can actually run and interpret linters, SCA, SAST, and DAST on its own.
The offensive side: Where AI-assisted penetration testing fits in right now.
Reality check: The actual limitations, failure modes, and logic gaps we run into with current AI security tooling.
Péter Tallosy - CTO @ BadCompany | Builder at mesh. from Badcompany
Petya is the CTO of BadCompany, an security startup. My background is mainly in physics and math. These days these are I also worked on various mathematical problems, did experimental research in nanophysics and also trained a lot of machine learning models. One year ago my team (Janos, Grego and Koppany) started working on (cyber)security problems and pentesting related to LLMs and basic agents. Then we continued with doing security audits for companies and developed a security solution for agents.
------------------------------------------------------------------
I will talk about current security related problems that need to be solved related to agents (for having good sleep at night) and how we are solving these problems: I'm Petya and I'm responsibility for technology in BadCompany, where we build security solutions for agents. We focus mainly on two directions: (1) robust and scalable policy engine, (2) robust and fast enforcement layer on the kernel.
Regarding capacity, we will start the guest list with 75 seats and increase the limit according to your level of interest.
Important Information
Registration: Claude Code Community reserves the right to approve or reject any registration. Please ensure your information is accurate, as incomplete or fake profiles will be declined.
Capacity: To ensure everyone's safety, we reserve the right to deny entry at the door if the venue reaches its maximum capacity.
Thank you and feel free to contact me, my DMs are always open!
tamas