Google, IBM & Meta Certificates — All 10,000+ Courses at 40% Off
One annual plan covers every course and certificate on Coursera. 40% off for a limited time.
Get Full Access
Explore a fundamental shift in AI agent construction through this 17-minute conference talk that challenges conventional scaffolding approaches. Learn why traditional methods relying on diagrams, algorithms, and handcrafted heuristics may be limiting AI agent development, and discover how embracing the "bitter lesson" from AI research history can lead to more effective solutions. Examine the tension between manual design approaches and generic methods that leverage massive computation, and understand why general-purpose learning often outperforms tailored engineering. Gain actionable strategies for designing robust, scalable AI agent systems that minimize over-engineering while maximizing the power of computational approaches. The presentation draws from decades of AI research insights to provide practical guidance on building more effective AI agents through rethinking fundamental scaffolding strategies, delivered by an industry expert with extensive experience in applied AI and customer support automation platforms.
