Anil Varanasi went to China expecting a short stay. A month or two, he and his brother Sunil thought, would help them learn how to make networking hardware. They stayed for more than a year. A return ticket is a modest thing to sacrifice to a company; an entire calendar requires a rather different temperament.
The company was Meter, founded in 2015. Before a product was ready to sell, more than four years would pass. Anil and Sunil had chosen an assignment with stubborn physical requirements. Somewhere between an idea about better internet and a functioning office network, somebody had to learn how the boxes were made.
A trip with no sensible return date
The brothers had already practised making ambitious things together: films, with particular ideas about the shots and music. Business had its own place in their upbringing. Their father ran a public company, and company filings featured in family conversation. The dinner-table reading included 8-Ks and 10-Ks. Other families presumably had lighter material.
Anil went on to study economics and computer networking at George Mason University. The pairing is useful shorthand for the work that followed. Networking asks how information gets somewhere. Economics asks who has a reason to help it arrive, what they should charge, and what happens when the arrangement goes wrong.
The invoice is an engineering decision
By March 2026, as Meter launched in the UK, Anil could describe the arrangement in compact terms. The same company builds the hardware and software, designs and deploys the network, and maintains it. Meter retains the equipment and charges a monthly fee based on square footage. It takes responsibility for hardware and installation costs.
That places an economic decision inside an engineering problem. If the customer is subscribing to a working network, shipping another device does not complete the promise. The design, the installation, and the subsequent support have to agree with one another. A business can have an immaculate product presentation and still leave somebody wondering whose number to call.
“We don’t even sell the hardware.”
Anil Varanasi, March 2026
The attraction of this model is easy to understand from a tenant’s end of the cable. Opening a space already requires enough negotiations. The network adds equipment, configuration, and maintenance to the list. Anil’s proposition puts those tasks into a continuing relationship. The bill arrives regularly; so does the obligation to make the arrangement useful.
The technical ownership behind that promise is unusually extensive. Meter developed its own network operating system and networking stack, using commercially available chips. Its integrated equipment reaches beyond switches and wireless access points to power distribution and security. A common management layer connects the pieces. Customers retain detailed technical controls, even though Meter owns the equipment.
Here, the practical virtue of the economics-and-networking education becomes visible. A recurring charge is a commercial choice. A shared operating system is a technical choice. Together, they determine how much of a customer’s problem the company can reasonably accept. Each decision creates obligations for the other. The invoice and the circuit board belong in the same conversation.
One continuing service, from equipment to everyday operation.
A reading list with room for a city
Anil’s personal website offers a wider view of what interests him. One essay concerns agglomeration: the benefits that arise when people and their work gather in the same place. He moves between historical clusters in Florence, Edinburgh, Vienna, Calcutta, and Berlin, then turns to research about productivity, wages, and the movement of knowledge.
His question is why more effort does not go into creating new clusters. The essay also points toward an underexplored subject: how proximity works inside a company. Read beside Meter, it suggests a recurring interest in the conditions that allow people to accomplish things together. The connection is an interpretation, but a fairly tangible one. Buildings contain people; networks connect their work.
Another piece, on prolific people, collects examples across music, invention, painting, and literature. Bach appears near Edison; Morricone shares a list with Danielle Steel. It is an agreeably untidy dinner party. Anil is interested in the sheer amount some individuals produce, and whether ordinary expectations of output account for them.
The list is useful as a portrait of attention. It puts technical invention alongside creative work, with neither asked to surrender its place. For a founder who once made films with his brother, that breadth makes sense. The question running underneath is about making: how much someone can produce, and what assumptions might be preventing everyone else from trying.
His essay on age approaches opportunity from another direction. It assembles examples of influence and achievement arriving later across different fields, and asks which generations have agency. He treats the pattern as a structural question. A patient company horizon and an interest in early opportunity can sit together quite comfortably: starting sooner gives a person more time to build.
Then there is his collection of things people still struggle to explain fully: flight, fire, magnetism, gravity, dark matter. It is presented as an invitation for better explanations. These are subjects familiar enough to pass without comment in daily life, yet capable of opening into difficult questions. Familiarity, in this reading, is a poor reason to stop investigating.
An audience of one can be enough
That interest in opportunity has taken a personal form. Anil and Sunil have described finding work online by people with small audiences: a video, a blog, a podcast, a software repository. When something catches their attention, they contact its maker and ask what financial support would allow six months of full-time work.
The mechanism is disarmingly direct. It begins with something already made and the possibility of more. There is a connection here to his interest in prolific creators, though the result of any particular offer remains its own story. A person with a small audience can still have the next piece of work waiting, provided somebody helps make room for it.
There is also a pleasing fit with his day job. The internet makes it possible to encounter a stranger’s work without sharing a city or an institution. A networking business operates underneath those encounters. Supporting a creator happens further along the same chain. One concerns access to an audience; the other concerns the ability to keep making something worth finding.
The network reaches the building
Meter’s more recent developments show how far that underlying infrastructure assignment has expanded. In February 2025, Anil announced a multi-year partnership with Microsoft. The stated work included networking development, Azure computing resources for custom models, and new distribution through Azure Marketplace. Command, Meter’s generative interface for network management, was part of the planned move onto Azure.
For the CEO, this adds another layer to the task. The company’s own equipment still matters, but so do the computing resources and commercial routes that let other organizations adopt it. Infrastructure has neighbours. Making it easier to buy can be consequential alongside making it easier to operate.
Meter’s Series C, announced June 12, 2025.
Led by General Catalyst.
That June, Meter announced $170 million in financing led by Hemant Taneja at General Catalyst. New and existing investors included Microsoft, Sequoia Capital, Baillie Gifford, and J.P. Morgan. The company named Bridgewater, Lyft, and Reddit among the businesses relying on its networks. The early manufacturing assignment now had a larger institutional audience.
In March 2026, Meter announced that it had acquired WiredScore. The acquired business had spent more than a decade assessing connectivity in buildings. Its relationships and standards addressed a decision that happens before an employee joins a wireless network: whether a space has the infrastructure a tenant needs.
The companies said WiredScore and SmartScore certifications would continue as independent standards. They also announced a London office as part of Meter’s European expansion. Anil’s ambition was becoming more closely tied to real estate: a working network as an expected feature of a building. The closer the promise gets to the lease, the earlier someone has to take responsibility for delivering it.
In May, Meter committed $100 million to a Partner Growth Fund. Qualified partners could receive support for customer migrations, training, marketing, development, and other work involved in replacing older networks. Building a product and building a route into the market require different kinds of effort. By this stage, the company was putting money behind both.
Someone still has to learn the craft
The next small detail is worth keeping beside those large numbers. In October 2026, Meter introduced grants of up to $2,000 for people starting in networking in the United States. Certifications, courses, study materials, and lab equipment were eligible uses. Applicants did not need an existing technology job or a degree.
A company can plan for decades, but someone still needs a first set of equipment to practise on. The grants recognise that practical threshold. They also bring the story back toward the brothers’ extended trip: learning how the machinery works takes time, access, and a willingness to begin before you know enough.
The underlying ambition is wonderfully ordinary. Enter a building. Connect. Work on whatever brought you there. For Anil Varanasi, making that sequence uneventful has required a remarkably eventful working life. The customer is entitled to miss the drama. After all those years spent learning how to build the boxes, a quiet day on the network would be a satisfying review.