Lunch & Hack Online Series: Decision-Making & Guardrails. Building Reliable AI Agents. (w/ The AI Collective)
Virtual Lunch & Hack
Join us for a Virtual Lunch & Hack, a biweekly online event where builders come together to learn something new, hear from great speakers, and hack on a fun project in real time.
Each session features a different topic. We’ll kick things off with a short talk from our speakers, followed by a hands-on hacking session where everyone gets a chance to build. At the end, selected participants can show what they made.
For this session, we’ll learn about Decision-Making & Guardrails: Building Reliable AI Agents.
As AI agents take on increasingly complex tasks, reliability becomes just as important as intelligence. In this Lunch & Hack, we'll explore how to build AI agents that make better decisions, respect guardrails, and behave predictably in production. We'll cover practical techniques for memory, planning, evaluations, and guardrails, followed by a hands-on hack where you'll build and test a reliable AI agent yourself.
For the hack part we'll use Runtime to build a reliable AI agent live. The challenge is to automate the biggest source of your day-to-day busy work using skills, guardrails, and evals to keep the agent's behavior predictable in production. Attendees will leave with a working agent and a practical framework for adding reliability checks to any agent they build going forward.
Check our agenda, future topics, past recordings or propose new topics at: RoryPlans.ai/Lunch
Speakers:
Carlos Volante - CTO at Runtime
Magda Jary - Researcher, ex Google
Adelina Martiniuc - CEO RoryPlans.ai
Format
20 minutes — Topic discussion with the Speakers
30 minutes — Group hacking session + Q&A
Show & tell — Share what you built
Carlos Volante
I'm Carlos Volante, co-founder and CTO of Runtime. I previously built financial and agent infrastructure at Modern Treasury and Finix, and before that worked on streaming infrastructure at Hulu.
Magda Jary
Magda Jary is a researcher and expert in digital transformation, AI literacy, and organizational leadership. With 16 years at Google driving go-to-market strategies, change management, and product adoption across Poland, London, and California, she currently explores the impact of AI on strategic decision-making as a PhD candidate at the University of Tennessee at Chattanooga. A certified mindfulness meditation teacher and 2023 Top Woman in Cloud recipient, Magda brings tech industry experience and academic inquiry together to help organizations navigate AI adoption with human-centered guardrails.
Adelina Martiniuc
CEO of RoryPlans.ai and the creator of RoBrain.dev, Adelina has over 15 years of experience in tech, including leadership roles at Vercel, Cruise, and Google. She wrote her thesis on AI planning and optimization 15 years ago, exploring agent-based systems long before they were widely called “agents.” She is currently building Rory, a multi-agent orchestrator that aligns human teams and AI agents in a single system of execution. RoBrain is a shared memory across your team and your AI agents, with judgment about what's worth keeping.
The AI Collective is a global non-profit building the human layer for the AI era. We unite 200,000+ leaders, builders, and stakeholders across 100+ forums worldwide to democratize the frontier, build trust, and coordinate how society navigates the rapid acceleration of technological progress.
All attendees and organizers at events affiliated with The AI Collective agree to our privacy policy and are subject to our code of conduct.