Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Watch this 57-minute conference talk exploring the practical realities of building production-ready AI agents at scale. Learn from Alex Salazar, CEO and Co-Founder of Arcade.dev, and Chiara Caratelli, Data Scientist at Prosus Group, as they share real-world insights on tool definitions, user experience design, and why agent autonomy may be overrated when millions of users are involved. Discover key lessons from iFood's implementation, understand the critical differences between API and agent intentions, and explore strategies for tool definition clarity and preemptive context loading. Examine how to effectively contextualize agent data, avoid prompt bloat in payment systems, and navigate the evolution of agent building practices. Gain insights into agent program scalability challenges and understand why multi-agent approaches may be reaching a dead end, while learning practical approaches to making AI agents work reliably in production environments with real users and complex automation requirements.
