"2025 has undeniably been the year of AI agents. Coming into the year, it was nearly impossible to tune into any conversation in the tech ecosystem without hearing the agent buzzword thrown around. In an attempt to gain my own understanding of this new space, I started reading up on AI, machine learning, and agents. This led me to Mastra, an open-source framework for building AI agents. The best part was that it's built in my go-to language, TypeScript."
"Not long after I started reading through the Mastra documentation, the creators announced mastra.build, an AI hackathon focused on creating pre-built Mastra projects to be used as a starting point for common agentic use cases. Perfect timing, honestly. So, I registered for the hackathon to put myself to the test. The result? An AI agent that lets you ask questions about a collection of video transcripts in natural language - and it even ended up winning a raffle prize too."
2025 saw broad adoption and interest in AI agents across the tech ecosystem. Mastra is an open-source framework, implemented in TypeScript, for building AI agents. The mastra.build hackathon invited developers to create pre-built Mastra projects to seed common agentic use cases within a two-week timeframe. Prizes recognized categories such as Best Use of Tools and Best Use of Auth. A Mastra product leader requested an agent to streamline video production by analyzing channel video transcripts and answering natural-language questions. A submitted template implemented an agent that queries collections of video transcripts and supports content analysis and production workflows. The template demonstrated practical integration of agent tooling, authentication, and transcript processing.
Read at LogRocket Blog
Unable to calculate read time
Collection
[
|
...
]