2 Sources
[1]
AgentMail raises $6M to build an email service for AI agents | TechCrunch
Just a couple of years ago, AI agents were mostly chatbots that could use basic tools. People were curious, but given concerns around reliability and security, as well as cost, the tech remained in the realm of early adopters. How things have changed. Coding agents like Claude Code, Codex, and
[2]
AgentMail raises $6M to give AI agents their own email inboxes
The San Francisco startup, backed by General Catalyst and Y Combinator, is betting that email, not some new identity protocol, is how AI agents will establish themselves on the internet. AI agents can already book your meetings, negotiate contracts, and handle support queues. What they have not
Share
Copy Link
San Francisco startup AgentMail has raised $6 million in seed funding led by General Catalyst to provide AI agents with their own email inboxes. The API platform enables two-way communication, threading, and labeling, positioning email as the identity layer for AI agents rather than building new protocols from scratch.
AgentMail, a San Francisco startup that emerged from Y Combinator's Summer 2025 batch, has raised $6 million seed funding to build an email service designed specifically for AI agents
1
2
. The round was led by General Catalyst, with participation from Y Combinator and Phosphor Capital2
. Notable angel investors include Paul Graham, Dharmesh Shah (CTO of HubSpot), Paul Copplestone (CEO of Supabase), and Karim Atiyeh (CTO of Ramp)1
.
Source: The Next Web
The API platform gives AI agents their own fully functional AI agents email inboxes, complete with real addresses capable of two-way communication for AI agents, threading, labeling, searching, replying, and parsing structured data from incoming messages
2
. A single API call creates an inbox without OAuth flows or manual setup2
.Co-founder and CEO Haakam Aujla, a former quantitative researcher at Optiver, founded the company alongside Michael Kim and Adi Singh with a clear thesis: email already serves as the identity layer for AI agents across the internet
2
. Rather than building an entirely new identity protocol, AgentMail leverages what already works for humans. "You give an agent an email address, and it can now use essentially any software service that already exists," Aujla explained1
.The platform was built from the ground up to provide AI agents a similar inbox experience as services like Gmail or Outlook, except without UI elements humans need
1
. "They shouldn't have to click buttons on a screen, because that's pretty clunky for agents to do. They should just be able to make API calls," Aujla told TechCrunch1
.Alongside the funding announcement, AgentMail launched an onboarding API that allows AI agents to sign themselves up directly, navigating to the platform and creating an inbox without any developer involvement
1
2
. The company revealed an unexpected development: "Autonomous agents have started signing up for AgentMail on their own, finding us through web search, navigating to our site, and creating their own inboxes without a developer in the loop," according to their launch post2
.The platform integrates seamlessly with LangChain, LlamaIndex, and CrewAI, working with any framework capable of making an API call
2
. This positions AgentMail as infrastructure for high-volume agent use cases that legacy providers struggle to support.Related Stories
AgentMail launched in August 2025 and initially focused on B2B use cases for companies needing to scale email communications
1
. Progress was incremental until OpenClaw's blockbuster debut in late January 2026 created immediate demand for agent infrastructure1
. User numbers tripled during OpenClaw's breakout week and quadrupled in February1
. The company now counts tens of thousands of human users, hundreds of thousands of agent users, and more than 500 B2B customers, according to Aujla1
.Use cases span supply chain coordination, loan collection, and procurement negotiations—all scenarios where traditional email providers like Gmail impose rate limits that make them unworkable for agent deployments at scale
2
. AgentMail provides a generous free tier alongside paid plans and enterprise subscriptions1
.Giving email inboxes to AI agents raises obvious concerns about potential misuse. Aujla outlined several safeguards: agent inboxes can only send 10 emails per day unless authenticated by a person; the platform imposes rate limits when detecting unusual activity; bounce rates are monitored; and new accounts are randomly sampled to filter for sensitive keywords
1
2
. Whether these controls prove sufficient as the platform scales remains a critical question.General Catalyst's Yuri Sagalov framed the investment around a fundamental shift: "Email is the heart of identity on the internet. Traditional identity services were not built with agentic use cases in mind, and AgentMail is building that part of the stack, starting with email," Sagalov said
2
. As AI agents take on more tasks currently handled by humans, they'll need credentials, reputation, and trust—the fuller architecture of online identity that AgentMail aims to provide.Summarized by
Navi
[2]