Jan 22, 2025

6:00 AM PST · January 22, 2025 Since it was founded in 2014, India-based Postman has made a name for itself as one of the most popular platforms for building and using APIs, with 500,000 organizations now using the service. Like with so many other SaaS services, though, its valuation today is reportedly down from its $5.6 billion high in 2021. But the company is now jumping head-first into the AI space with the launch of an AI agent builder that combines large language models and the Postman API platform, as well as a visual editor that brings the two together to help even non-developers build and test AI agents. At first, jumping into AI agents may seem like an odd choice for a company like Postman, but as Postman CEO and co-founder Abhinav Asthana noted when I talked to him ahead of the agent builder’s launch, to be useful, these agents have to interact with a wide variety of services — and the way to do that is through APIs. “Our customers actually started asking us about this. ‘Okay, this agent thing is kind of coming up. What are you guys doing about it?’ And we looked at the patterns in which they are stringing this together. Whether they’re using homegrown systems or they’re using some third-party stuff, APIs are a key part of it, because there are two kinds of APIs: one is the LLM itself, like the OpenAI API, or the Anthropic’s API, or maybe a homegrown API. And the second bit are tools the agent would use. So we were like, okay, Postman is helping developers do all things APIs and if this is emerging as a pattern, we should do something here,” explained Asthana. Postman agent workflows.Image Credits:Postman He also noted that a growing number of Postman users are non-developers, who want to use the service for building simple, API-centric applications. “So when AI technology started emerging, we had to test it out a lot. We saw like, ‘okay, can this be used?’ And it looks like no-code, plus AI, plus agents is a good combo to bring together,” Asthana said. Postman’s API network already features an API hub with contributions from over 18,000 companies, which now, in turn, also allows its users to build AI agents that can interact with these APIs. To ensure that these agents work as expected, Postman is adding testing and evaluation tools to the service to test prompts and inputs for the models it supports (think OpenAI, Google’s Gemini, Anthropic’s Claude, Cohere and Meta’s Llama, for example). It then allows users to use Postman Flows, the company’s visual development environment, to build these agents and have them interact with APIs. Postman LLM evaluation.Image Credits:Postman Right now, Postman’s AI agents are mostly focused on enabling back-end operations. While developers could use the service to power front-end services, Postman itself doesn’t offer any user interface for directly interacting with these agents yet. Over time, this may change. “We are all very bullish on agent software, but I think the modalities in which we interact with these systems is also going to evolve,” Asthana said. Postman, of course, is not the only company in this space, but on top of having its existing API hub, Asthana also believes that it can differentiate its service from others — and especially the large cloud providers — by offering a better developer experience. “In my opinion, the cloud providers always prioritize consumption of the cloud over developer experience, and I think they’re willing to partner with anyone who does actually give the right developer experience,” he said. The company is still figuring out how to best charge for this new service. Pricing, Asthana said, will evolve over time as the team works out how to best determine the value that these agents bring. Topics