GenAI Cracow #21

Most of us (if not all already) can build generative ai systems in python or typescript. What about C#, Ruby and other programming languages?

Doors open: 5:30 PM

​

Abstract

​The Evolving Agent in C#

This presentation offers a hands-on guide to building a self-learning LLM-based agent using .NET, C#, and Semantic Kernel. The session showcases (through live coding demonstration) factual learning where the agent expands its knowledge base in real-time, behavioral learning where it adapts to user feedback and preferences, and capability learning where the agent writes and integrates its own new functions. The presentation combines theoretical foundations with live demonstrations, showing how modern AI agents can achieve practical "self-learning" capabilities within the .NET ecosystem.

​Multi-Agent architecture with Ruby & the AI open-source ecosystem

Ruby lacks mature AI libraries, and most available gems are just Python/JS ports that feel awkward. This talk covers building multi-agent systems in Ruby by skipping frameworks entirely and using the OpenAI/Anthropic API directly. I'll show how we handle tools and MCP protocol, manage context between agents, and optimize for speed and cost. We'll also discuss what makes Ruby different in this space and why sometimes having fewer options forces better architectural decisions.

​

Speakers

​Tomasz Kaszta

Experienced Software Engineer with 20+ years in full-stack development, system architecture, and startup leadership, now focused on solving business problems with generative AI and LLMs. Proven success in designing and delivering scalable, secure, high-performance systems for top financial institutions in London. Skilled in providing technical leadership and driving innovation across greenfield and modernization projects.

​Szymon Kurcab

Head of AI Labs @ Tropic, transforming complex spend data into practical AI assistants for procurement teams. Recently created the ruby-mcp-client gem that allows MCP support to Ruby for developers. Former CTO at Chargify (now Maxio). He co-created repertuary.pl. Actively contributes to the Ruby community through KRUG meetups

Linkedin | Github

​

Sponsors

​EPAM Systems

EPAM Systems is a leading provider of digital transformation services in the world and a leader in the development of digital and physical products and digital platform engineering services. Through its innovative strategy; integrated advisory, consulting, and design capabilities; and unique 'Engineering DNA,' EPAM's globally deployed hybrid teams help make the future real for clients and communities around the world by powering better enterprise, education and health platforms that connect people, optimize experiences, and improve people's lives.