Portia AI, a London-based startup co-founded by former leaders from Stripe and Google, has raised £4.4 million to help developers deploy AI agents with real-time oversight, security, and auditability.
The round was led by General Catalyst, with backing to accelerate the development of Portia’s open-source framework and cloud platform for building secure, production-ready AI agents.
As AI agents gain traction for their autonomous capabilities and LLM-powered reasoning, many pilots struggle to move beyond testing. The problem? Lack of visibility, control, and trust. Portia AI addresses these gaps—particularly important in regulated sectors like finance, where compliance and accountability are critical.
“Developers want AI agents that act like reliable collaborators, not unpredictable black boxes,” said Mounir Mouawad, Portia AI’s co-founder and CEO.
Designed for Production, Built for Trust
Portia AI emerged from stealth earlier this year, co-founded by Emma Burrows (former CTO of Stripe UK and tech lead at Charlotte Tilbury) and Mounir Mouawad (who helped launch Google Pay across 30 markets and led product growth at Stripe). The duo brings a rare blend of deep technical expertise and product leadership, with a shared vision to make agent-based automation controllable, predictable, and secure.
Their open-source framework empowers developers to:
- Create step-by-step, auditable agent plans
- Pause execution to clarify missing inputs or secure human approvals
- Authenticate agents securely and just-in-time to external tools and APIs
This means developers and teams can monitor, intervene, and customise agent actions with precision—while ensuring regulatory standards and internal compliance are met.
“We’re making AI agents safe and reliable to use in real business processes—without needing to be an AI expert,” added Emma Burrows, Portia AI’s co-founder and CTO.
Advertisement
Built for Developers, Backed by AI Expertise
Portia’s platform includes:
- A just-in-time authentication layer for secure tool access
- A growing catalogue of pre-integrated apps like Google, Slack, GitHub, and Zendesk
- An SDK and cloud dashboard to customise behaviours and scale use cases
- Full logging and serialization of agent plans for auditing and compliance
From fintech co-pilots to customer service bots, Portia’s framework is designed for flexibility—giving developers the tools to launch intelligent agents across high-stakes workflows.
Inspired by Sci-Fi, Driven by Real-World Problems
The name “Portia” pays homage to a spider protagonist in Adrian Tchaikovsky’s Children of Time—a species that builds an advanced civilisation through learning and collaboration. It reflects the team’s mission: to build AI that enhances, rather than replaces, human decision-making.
“Our goal is to empower developers everywhere to deploy AI agents with confidence, transparency, and safety,” said the founders.
Backers See Portia as a Foundation Layer for AI Agent Infrastructure
Juliet Bailin, Partner at General Catalyst, called the company’s timing and leadership “brilliantly positioned”:
“Portia AI is doing the hard work of making agents not only autonomous but accountable. Emma and Mounir’s track record speaks volumes—they’re building infrastructure that could define the next decade of intelligent automation.”
Portia AI is hiring and expanding its developer team, with multiple new features and integrations slated for release in 2025. As the AI agent ecosystem continues to explode, Portia is carving a niche as the backbone of secure, production-grade agent deployments.
