

Build the Harness: A Hands-On Challenge in AI Agent Reliability - Workshop at Compass AI & Tech Summit
A prompt and a model alone don't make an agent you can rely on in a company. What makes it accurate, secure, and trustworthy is the harness around it and getting that right takes more decisions than most teams expect.
In this hands-on workshop, you will compete solo or as a team to tackle example tasks from a corporate environment. You’ll build an agent to solve these challenges using real AI models on a set token budget. Every attempt is scored live, and the highest score wins.
After the competition, we’ll look behind the scoreboard to analyze how the agents performed on unseen inputs, which approaches held up, which ones broke, and what a genuinely strong solution looks like. You’ll leave with a practical understanding of what it takes to build AI agents that work reliably in a real business setting.
Prerequisites & What to Bring
Bring a laptop (at least one per team).
Open to all experience levels, so no coding background is necessary.
About the Host
Bálint Horváth — Head of Product & Innovation @ LogiNet | Co-Founder @ StickyPrompts
Bálint leads product strategy, manages AI projects, and oversees technical solutions at LogiNet, while driving practical AI adoption across teams and organizations. Starting his coding journey at age nine, his deep technical foundation spans over a decade in eCommerce, SaaS, and IT leadership, giving him a complete end-to-end view of software from the first line of code to the final customer. Outside of building marketable products, Bálint does improv, mentors startups, and explores hard tech (though you won't catch him running AI agents just to keep them busy).
Please take under consideration your application, as the places of the workshop are limited. If you cannot make it to come, please select the 'not going' option, so that others can join.