To framework or not to framework? That is the question.

To framework or not to framework” explores the tension every developer feels between building everything from scratch and standing on the shoulders of tools, libraries, and frameworks created by others. In a world where the raw APIs of OpenAI, Anthropic, and other LLM providers are just a curl command away, it’s tempting - and even fun - to roll your own abstractions. But as our apps grow, so do the trade-offs: complexity, duplication, maintenance, and the painful rediscovery of known problems. In this talk, we’ll unpack how leveraging frameworks like Mastra, ADK, and others doesn’t make you less of a builder - it makes you faster, more durable, and more focused on what truly matters: delivering value, not reinventing wheels. When the landscape shifts as fast as AI, choosing not to build everything yourself isn’t a shortcut - it’s a strategy.