Case Study: PLAID boosts AI agent development productivity with Mastra

Why PLAID Japan builds agents on their Google Cloud infrastructure with Mastra

PLAID Japan, the developer of the KARTE customer engagement platform, faced challenges with GUI-based AI development tools like Dify and n8n, which disrupted their engineering workflows and lacked proper TypeScript support. Their team needed to build sophisticated AI features but was constrained by tools that were not code-first. They turned to the vendor Mastra for a solution built around its TypeScript framework and tools.

By implementing Mastra on their existing Google Cloud infrastructure, PLAID's engineering team gained a code-first solution that improved collaboration and development productivity. The Mastra framework allowed them to easily build agents, integrate with external tools like Slack via MCP, and use the playground for prompt adjustment. This shift from GUI tools to Mastra has made AI agents more accessible for internal use and is now enabling the development of advanced customer-facing agents.

PLAID

Takanori Koga

Engineer