Mastra, an open-source TypeScript framework designed to help developers build and deploy artificial intelligence agents, has raised $22 million in Series A funding to expand its platform and accelerate its mission of making AI agent development accessible to more software developers. The round was led by Spark Capital, bringing Mastra’s total funding raised to $35 million.

The Series A investment follows Mastra’s earlier $13 million seed round, which was backed by a group of investors including Y Combinator, Gradient Ventures, basecase, and several technology executives and angel investors. The company said the seed financing came from more than 100 investors as it worked to develop tools for building production-ready AI agents.

Founded by Sam Bhagwat, Shane Thomas, and Abhi Aiyer, Mastra was created by the team behind Gatsby, the open-source React framework used by developers to build modern web applications. The founders began working on Mastra after identifying challenges developers faced while building AI applications, including managing agent behavior, tool usage, memory, workflows, and evaluation systems.

Mastra provides an open-source TypeScript framework that enables developers to create AI-powered applications and agents. The platform includes tools for building agents, workflows, retrieval-augmented generation systems, memory features, and evaluation capabilities. It is designed to help developers move from AI prototypes to production applications by providing reusable infrastructure for complex agent-based software.

The newly raised capital will support Mastra’s continued product development, expansion of its engineering team, and growth of its platform for enterprises and developers. The company is focused on improving the tools needed to build, test, deploy, and manage AI agents as organizations increasingly adopt autonomous software systems.

Mastra’s funding comes as interest in AI agents continues to grow across the software industry. Unlike traditional software applications that follow fixed instructions, AI agents are designed to make decisions, use tools, manage workflows, and complete tasks with varying levels of autonomy. Developers building these systems require specialized frameworks that can handle reliability, monitoring, memory, and deployment challenges.

The company has positioned itself as infrastructure for this emerging category of software development. Mastra’s framework is built around TypeScript, a programming language widely used in web development, allowing developers familiar with modern JavaScript ecosystems to create AI agents without needing to rely exclusively on Python-based tools.

Investor support from Spark Capital highlights increasing venture interest in developer infrastructure companies focused on artificial intelligence. Spark Capital has invested in technology companies across software, consumer technology, and enterprise platforms, and its leadership of Mastra’s Series A reflects confidence in the growing market for AI development tools.

Earlier backing from Y Combinator helped Mastra accelerate its early development. The company participated in Y Combinator’s Winter 2025 batch, joining a group of startups supported by the accelerator’s global founder network.

Gradient Ventures also participated in Mastra’s seed financing. The Google-affiliated venture fund focuses on artificial intelligence startups and supports companies developing technologies that expand the capabilities and adoption of AI.

Mastra has gained adoption among companies building AI-powered applications. The company has highlighted usage by organizations including Replit, Brex, MongoDB, Workday, Salesforce, Sanity, SoftBank, and Marsh McLennan. The platform is designed to support a wide range of AI agent applications, from enterprise workflows to developer tools.

Alongside its funding announcement, Mastra introduced additional platform capabilities aimed at helping teams operate AI agents at scale. These include Mastra Studio for evaluations, logs, and traces; Mastra Server for deploying agents and workflows; and Memory Gateway for managing agent memory.

The company’s growth reflects a broader shift in software development as businesses move from experimenting with generative AI tools toward deploying autonomous systems capable of performing complex tasks. Developers increasingly need frameworks that provide structure, reliability, and visibility for AI-powered applications.

With its $22 million Series A led by Spark Capital and previous backing from investors including Y Combinator, Gradient Ventures, and basecase, Mastra plans to continue building infrastructure that enables developers to create and operate AI agents. The company aims to become a core platform for the next generation of AI-powered software development.

Related Articles