Linq announced a $90 million funding round this week, led by Andreessen Horowitz alongside TQ Ventures, Mucker Capital, Abstract, Anthos and Left Lane Capital. The capital will support the startup's expansion into voice and computer automation capabilities for AI agents. Ben Horowitz, co-founder and general partner at Andreessen Horowitz, is joining Linq's board.

The company addresses a fundamental limitation in how AI assistants currently operate. When you ask an AI to reschedule a dentist appointment, it typically drafts a message for you to send, then stops. Completing the task actually requires sending text, placing a call, or navigating a website—actions most AI systems cannot perform independently. Linq aims to bridge this gap by providing developers with the infrastructure to give their assistants these capabilities.

A phone for the assistant

Text-based personal assistants capable of taking actions on behalf of users represent one of the fastest-growing segments in AI development. The interface requires no learning curve, and requests can be expressed with natural specificity—"Friday after 3, preferably at the same office" conveys more nuance than traditional search filters and phone menus.

Linq identified an important dynamic: the person making the request and the business receiving it often prefer different communication channels. A user might prefer handling everything through text, while a receptionist may respond better to a phone call. Linq began by focusing on the messaging side of this equation. The startup reports processing more than 1 billion messages across customers including DoorDash, Town, Tomo, and hundreds of additional companies. DoorDash leverages Linq to enable customers to search for food, build orders, and complete purchases within a single message conversation.

The company's new voice offering, branded HD Calling for Agents, expands the toolkit available to developers. An assistant can receive instructions via text and switch to voice calling when that channel better serves the task. For users, the benefit is straightforward: they can resume other activities while awaiting a text confirmation of their new appointment time.

A computer to do the work

Many tasks demand that someone sit at a computer and navigate software interfaces. Linq's computer use product grants agents access to physical computing infrastructure, allowing them to interact with applications. This capability opens possibilities for developers building personal assistants, particularly when businesses operate websites that customers can use but lack application programming interfaces for automated access.

Clicking through five screens to make a payment will become two words to an agent: 'pay rent,' Linq CEO Elliott Potter said in the announcement. "We help developers build customer-facing agents that use human tools to do human work."

Elliott Potter, Linq CEO

The real power emerges when tasks flow across multiple channels. A receptionist might confirm a new appointment by phone, then send a link to a form requiring completion before the booking is finalized. A voice-only assistant would need to hand off the task at that point. An assistant with access to messaging and computer capabilities could complete the entire workflow independently.

Selling tools to the personal assistant boom

Linq's strategy positions it to sidestep one of the central uncertainties facing the personal AI assistant market: whether users will adopt a single general-purpose assistant handling everything from food delivery to rent payments, or rely on specialized agents provided by familiar businesses. Linq stands to benefit from either outcome, since both scenarios require developers to establish communication channels and execute tasks.

Agents will communicate and get work done just like people do. Linq gives agents the infrastructure to do both at scale.

Ben Horowitz

Should personal AI assistants gain widespread adoption, the smartphone could transform into a delegation device for errands that would otherwise consume a lunch break. A user texts their request, the assistant calls the office, navigates the booking system, and returns with a confirmed time. Linq's $90 million investment reflects confidence that as expectations around AI assistance grow, developers will increasingly require phone and computer infrastructure for each assistant they build.

Source: The Next Web