Parsec Founders Raise $8.75M From a16z to Give AI Agents Their Own Remote Workstations

Phaze Secures a16z Seed Funding for Shared Compute Infrastructure

Phaze, the new venture from Parsec founders, has raised $8.75 million in seed funding led by Andreessen Horowitz to address a critical inefficiency in enterprise computing1

1

2

2

. The round included participation from Index, NextView, Makers Fund, and Lerer Hippeau1

1

. Chief Executive Benjy Boxer, whose previous company Parsec Technologies was acquired by Unity for $320 million in 2021, is betting that companies waste significant resources on idle hardware1

1

. The Utah-based startup has developed a platform that transforms expensive workstations—often costing upwards of $10,000—into shared pools of elastic compute accessible to both human workers and AI agents2

2

.

Solving Dual Enterprise Challenges Through Remote Workstations

Phaze addresses two pressing problems facing modern enterprises. First, computer-use agents frequently hijack workers' screens while automating mundane business tasks, directly impacting human productivity2

2

. Second, many IT departments follow a policy of assigning one powerful workstation to each employee, meaning that expensive hardware capable of rendering complex 3D design work and compiling massive codebases sits idle during meetings, lunch breaks, or time off2

2

. Instead of buying each employee a top-end workstation, companies can pool machines and distribute them as needed to staff or AI agents1

1

. The platform uses proprietary low-latency streaming technology to deliver the same responsiveness and visual fidelity users would experience with local hardware2

2

.

AI Agents Remote Access Through Model Context Protocol

Phaze enables AI agents to control remote workstations through the Model Context Protocol, a common standard for linking AI to tools that the company still labels experimental1

1

. This specially developed MCP server allows computer-use agents to log into and operate expensive workstations the same way human employees would, navigating operating systems and launching applications directly rather than working through expensive and brittle APIs2

2

. Agents receive only the access permissions their users have, maintaining security protocols1

1

. Designers and engineers can run demanding software on high-performance compute resources while AI agents simultaneously control machines' screens and applications1

1

.

Human-Agent Collaboration for Complex Workflows

The platform supports human-agent collaboration, allowing both to work together on complex tasks2

2

. In one demonstration, a Slack bot handles IT support requests by logging into a user's computer to diagnose problems, then seamlessly passes the session to a human engineer when the solution requires more nuanced intervention than the agent is authorized to provide1

1

2

2

. Boxer emphasized this flexibility: "The computer sitting on your desk shouldn't determine what compute you can access. People need powerful workstations to do their jobs, and now AI agents need computers they can actually operate"1

1

2

2

.

Growing Market for Agent Infrastructure

Phaze enters a rapidly expanding market for agent infrastructure. Baseten acquired Blaxel, which builds sandboxes for AI agents, in September, while Perplexity can transform a Mac mini into an always-on agent1

1

. The demand appears substantial—AI agents helped sell out Apple's Mac mini in the US this spring1

1

. Andreessen Horowitz General Partner Martin Casado highlighted the market opportunity: "Phaze is turning those machines into shared infrastructure for people and AI agents. Benjy and the team have been building in this space for more than a decade and they bring that experience to a new era of remote access"2

2

. Phaze plans to use the funding to accelerate go-to-market efforts, onboard new customers, expand its engineering team, and develop additional features2

2

.