Helping Americans believe in government again

Section 1

The power of government

In the middle of the twentieth century, the United States built things that still seem improbable. The Manhattan Project coordinated 125,000 people in three years to split the atom. The Interstate Highway System connected every state in a decade. NASA put a man on the moon in 1969, and Medicare enrolled 19 million people in its first year.

These weren't flukes. They came from a country where citizens and institutions had a functioning relationship, where Americans believed the system mostly worked, politicians crossed the aisle, and infrastructure got built. That baseline has eroded. Pew Research Center polling shows trust in government has fallen from 77% in 1964 to under 20% today, a trend that has persisted through every administration for sixty years.

We believe this is one of the most important problems in the country. Solving it requires a fundamentally different approach to how technology is built for government.

Section 2

Why trust matters

Trust is the mechanism through which 330 million people agree to be governed by institutions they didn't individually choose and fund services they may never use. When it breaks down, voter turnout declines, tax compliance weakens, and fewer people are willing to serve in government, further degrading institutional quality and feeding the cycle.

Yet this is what most commentary on this subject misses: Americans don't build their views about government from cable news or social media. They build them from direct experience: the permit office, the paper form that could have been a web page, the inspection that takes six weeks to schedule. Each of these interactions reinforces the same conclusion that government is slow and inefficient. And that conclusion, once formed, doesn't stay contained to the agency that created it. It becomes a worldview about whether institutions are worth defending at all.

This problem isn't partisan. Ezra Klein and Derek Thompson argue in Abundance that the left must reckon with the role of process in America's inability to build. Elon Musk created DOGE to attack waste from the other direction.

Section 3

The technology gap

Over the past twenty years, the private sector has gone through one of the most dramatic technology transformations in human history. You can hail a car in 90 seconds, manage your finances without visiting a bank, and have a conversation with an AI that generates useful answers in real time.

Then you try to file a building permit.

In thousands of municipalities, the process hasn't changed in twenty years. A contractor who runs his entire business from an iPhone walks into a government office and is handed a clipboard. A resident who files federal taxes online in twenty minutes spends three hours on hold asking about a zoning variance. A business owner who launched on Shopify in an afternoon waits eleven weeks for her occupancy permit.

The gap between what Americans experience as consumers and what they experience as citizens grows every year. The permitting clerks, building inspectors, and planning directors we work with every day are dedicated professionals operating under real constraints. They care deeply about their communities. The problem is the technology they're forced to use.

Section 4

Government cannot fix this alone

If the technology gap between the public and private sectors continues to widen, public confidence will continue to fall. But government is structurally unable to close this gap on its own.

Local governments cannot hire software engineers at competitive salaries. The hiring process itself is slow, often requiring months of postings, reviews, and approvals before an offer can be extended. Government institutions are optimized for process compliance, risk avoidance, and oversight, all features of public accountability that also make it nearly impossible to compete for engineering talent.

Government needs product companies that build software the way the best technology companies do, that attract the talent government cannot hire, and that are accountable for delivering results because their survival depends on it.

Section 5

How AI changes the equation

For years, modernizing government technology meant playing catch-up. Digitize the forms. Put the portal online. Make the website mobile-friendly. AI changes the equation in two ways.

First, it lets small engineering teams build complex software at a pace that would have required far larger teams just a few years ago. Government technology has historically required either a generic product that fit nobody well or an army of implementation consultants to customize it. AI collapses that tradeoff, making it possible to build software that is deeply configurable and still economically viable across thousands of agencies with different codes, workflows, and requirements.

Second, AI does actual work. It can review permit applications against local code, answer resident questions at any hour, flag violations in submitted plans, route inspections by priority and geography, and generate the reports that consume hours of staff time every week in every local government in the country. Government work is rules-based and context-structured, exactly the conditions where AI performs most reliably.

Take Nederland, Texas, a city of 18,000 where a single code enforcement officer handled every complaint, inspection, and notice entirely by hand. During peak seasons he received twenty-five complaints a day but could only work through nine. After deploying GovWell, daily capacity increased by 65%. They processed nearly 900 violations in eleven months, and their officer now spends 80% of his day in the field with residents instead of at a desk doing paperwork.

The vision, when this works at scale, is not abstract. We know what it looks like. It’s a resident submitting a permit application online in seconds and getting an AI-assisted review within minutes or a contractor checking the status of his inspection from his phone while driving to a job site. It’s a building department that used to spend 70% of its time on data entry now spending that time on the work that requires human judgment: reviewing complex plans, meeting with residents, and advising developers.

Section 6

This is the GovWell moment

GovWell is a technology company focused entirely on local government, built and staffed by engineers and designers who could work anywhere in tech but choose to work here. Our approach is simple: hire people who are among the best at what they do, focus them on a public sector problem, build something great, and earn the right to take on harder problems over time.

We are currently focused on AI-powered permitting and licensing for local governments. We chose this starting point deliberately. Every business in the United States needs permits. When a contractor can't get a permit for three months, that's a real person not earning income. When a homeowner can't get an inspection scheduled, that's a family unable to move into their home.

GovWell today powers workflows for municipalities and counties in more than 30 states, serving millions of residents. Customers report processing time reductions of up to 90%.

But permitting is where we start, not where we finish. As we earn the trust of the agencies we serve, we'll take on harder problems: statewide systems that unify fragmented local processes, unemployment benefits administration where delays mean families going without income, and visa processing where backlogs affect millions of people trying to build lives in this country.

Our incentives are aligned with the public interest. If our software makes a building department faster, more transparent, and less expensive to operate, we grow. If it doesn't, we lose the contract. We don't succeed unless the public servants who use our products succeed, and they don't succeed unless the residents and businesses they serve have a better experience than before.

Section 7

It’s time to believe in government again

GovWell alone won't restore trust in American government. The problem is too large for any single company, and we know that. What we can do is prove the model works.

The civic technology sector today is dramatically underfunded relative to its importance. Some of the most capable engineers in the country spend their careers optimizing ad click-through rates and refining recommendation algorithms. If even a fraction of that talent were redirected toward public institutions, the compounding effect over a decade would reshape what government is capable of delivering. We want GovWell to be the proof point that triggers that shift.

In 1962, President John F. Kennedy told an audience at Rice University why America was going to the moon. "We choose to go to the moon in this decade and do the other things," he said, "not because they are easy, but because they are hard." President Kennedy was talking about space, but he was also talking about what this country is capable of when it decides something matters enough to build.

We think government matters enough.