AI System Design: From Idea to Production - Apoorva Joshi, MongoDB
Building production AI systems requires a four-phase framework—requirements, system design, evaluation, optimization—not just vibe coding.
“Specs are the new code. The art is in defining the product requirements, the system design, and evaluate criteria so you can be confident that your AI coding buddies are building the right thing.”
A MongoDB developer advocate presents a repeatable four-phase framework (product requirements, system design, evaluation/monitoring, cost optimization) for taking AI systems from idea to production, illustrated with a health insurance claims review use case. The talk argues that vibe coding fails for high-stakes applications, echoing Anthropic and OpenAI messaging that specs and evals now matter more than code. It's a useful practitioner methodology but a conference how-to rather than a major industry announcement.