Everyone is watching the AI arms race in software. Foundation models. Copilots. Chat wrappers. Code generators. Harnesses. $700 billion in combined capex from Meta, Microsoft, Amazon, Google, and Oracle this year alone, almost all of it aimed at digital infrastructure.
Meanwhile, the largest sector of the American economy that actually builds physical things is collapsing under its own weight. And almost nobody in the AI world is paying attention.
Construction is a $2.2 trillion industry in the United States. It represents 4.4% of GDP. It employs 8.3 million people. And it is one of the only major sectors where productivity has actually declined over the past 40 years. McKinsey has called it one of the least digitized industries on the planet with it ranking just above agriculture and hunting.
This isn't a market that needs another project management dashboard. It needs a fundamental rearchitecting of how work gets scoped, sold, staffed, executed, invoiced, and collected. And that is exactly what RestoreFast is doing.
We are an AI-first restoration and construction contracting company operating primarily on U.S. military installations. We did 8 figures in revenue in 2025, with only ~10 employees. Our net income grew 180%.
We are in stealth, about to make our play. This is our thesis.
The Problem Nobody Wants to Talk About
The construction industry is facing a crisis that no amount of federal spending can fix. According to the Associated Builders and Contractors, the sector needs to attract roughly 349,000 net new workers in 2026 just to maintain current output, and that number is projected to climb to 456,000 in 2027 as spending growth resumes. The Associated General Contractors of America has reported that 92% of construction firms that are actively hiring have difficulty finding qualified workers.
This is not a cyclical downturn. It is structural decay.
Approximately 41% of the current construction workforce is expected to retire by 2031. One in five construction workers is already over 55. The pipeline of younger entrants is razor-thin, only 7% of potential job seekers even consider construction as a career. The National Association of Home Builders estimates that workforce constraints cost the U.S. economy $2.7 billion annually in project delays alone, and the skilled labor shortage costs the home building sector $10.8 billion per year between increased carrying costs and the roughly 19,000 homes that simply never get built.
And here's the part that should terrify every general contractor, facility owner, and government procurement officer in the country: the immigration enforcement environment in 2025–2026 has compounded the problem dramatically. Immigrants make up 34% of all construction workers nationally. In trades like drywall, roofing, and plastering, that share exceeds 60%. Nearly 28% of construction firms have reported workforce disruptions tied directly to enforcement activity. Residential construction timelines that once averaged six to eight months are now stretching to nine to twelve.
The industry's response? The same playbook it has run for 50 years: raise wages (construction wages are up over 4% year-over-year, with some firms increasing pay 20%+ just to compete), post on job boards, and hope someone shows up. Forty-five percent of firms are reporting project delays caused directly by worker shortages.
Warren Buffett built his career investing in boring, profitable businesses with clear competitive moats. He said publicly that the four largest companies in the world no longer need net tangible assets to produce earnings - that we have become an "asset-light economy." He was talking about Apple, Google, and their peers. But the principle applies just as powerfully in reverse: the industries that do require tangible assets and physical execution are the ones where AI-native operators can build the widest moats. Because the barrier to entry isn't code. It's concrete, compliance, boots on the ground, and the institutional knowledge to deliver.
Software can be commoditized overnight. A physical contracting operation running AI-native systems cannot be replicated by a competitor who hasn't already built the underlying business.
What RestoreFast Actually Is
RestoreFast is not a tech startup that talks about construction. We are a construction company that builds technology into every layer of how we operate.
Our roots are in residential losses. Think pipe bursts, flooding, fire damage in single-family homes. The kind of work you get by answering the phone at 2 a.m. and showing up before the water reaches the second floor. From there, we started knocking on doors at apartment complexes and commercial properties, building relationships with property managers and building owners, and earning repeat work through execution speed and reliability. Our board chairman brought over 30 years of construction experience, grey hairs, and a deep client network that opened the door to general construction projects, not just emergency response. That meant we were servicing buildings from the time they were in architectural planning, through active construction, through closeout, and into ongoing commercial maintenance. Homeowners. Property managers. Building owners. Developers. We had touched nearly every category of private-sector client our industry serves.
But we've been watching where the money is moving. Federal infrastructure spending was accelerating. Government-backed construction programs were creating multi-year project pipelines with predictable demand. We saw the tailwinds and made a deliberate decision to incorporate public-sector work into our service portfolio, and that bet has become our primary growth lever.
Today, we run restoration and construction on U.S. military bases - currently across 8 installations spanning Florida, Maryland, Texas, Georgia, Pennsylvania, North Carolina and Washington, D.C. Our anchor account in this portfolio has nearly quadrupled in project size, making it the largest contract in company history by 1 order of magnitude. And all of this happened in just 1 year.
What makes us structurally different is not the vertical. It is the operating model.
The average construction company generates somewhere between $148,000 and $420,000 in revenue per employee, depending on firm size. The cross-industry average across all sectors is approximately $350,000. RestoreFast generates north of $1 million per employee. That ratio is the output of a deliberate architectural decision to replace headcount with systems at every possible layer of the business.
When Elon acquired Twitter, one of the first things he flagged was revenue per employee, pointing out that Twitter's $625,000 per employee was anemic compared to Apple's $2.37 million or Google's $1.9 million. Investor Keith Rabois predicted that revenue per employee would become the defining metric as companies shifted from growth-at-all-costs to efficiency. Musk then proceeded to cut roughly 80% of Twitter's workforce and kept the platform running.
We are applying the same logic to construction, except we are not cutting an existing bloated workforce. We never hired one in the first place. We built the systems first and let the systems absorb the functions that a traditional contractor would staff with 50 to 100 people.
Every project we deliver is a training run. Every field event is logged into our fractal project tree structures: hierarchical data models that capture the full anatomy of a job from scope to closeout. Every escalation is handled gracefully by our proprietary agent runtime, which routes decisions, flags anomalies, and learns from the delta between what it suggested and what the operator accepted. The observation-to-improvement loop compresses with each cycle. The system does not just record what happened, it internalizes why it happened and adjusts. The company gets smarter with every single job.
The Financial Proof
This is not a pitch deck with projected numbers. These are actuals.
In 2025, RestoreFast's revenue experienced a 71% YoY increase. During that same period, salaries and wages decreased 17% decline. Read that combination carefully, because it is the signature of what is happening inside this company: revenue up, headcount cost down, simultaneously.
Operating income grew 143%, from $1.1 million to $2.7 million. Operating margin expanded from 18.9% to 26.7%, marking a 41% improvement. Net income nearly tripled, growing 180% YoY.
General and administrative costs as a percentage of revenue fell from 36.4% to 23.6%, a 35% compression. Total debt dropped 63%, from $1.17 million to $431,000. Revenue per dollar of property, plant, and equipment went from $22 to $337, a 15x increase.
We added $4.2 million in new revenue with negative headcount growth and negative asset growth.
These are not the financials of a construction company. They are the financials of a service company that knows how to use modern day software. And that distinction is the entire thesis.
Deloitte's 2026 Engineering and Construction Industry Outlook notes that the industry faces persistent inflation, elevated interest rates, tariff uncertainty, acute labor shortages, and tightened margins. Construction material costs hit a 40-year high effective tariff rate of 25% to 30% in 2025. Project abandonment activity surged 88% year-over-year. The industry is squeezing.
RestoreFast is expanding because of the squeeze, not in spite of it. Our model thrives in exactly the environment that breaks traditional contractors: when labor is scarce, materials are expensive, and margins are thin, the company that can execute with fewer people and better systems wins every time.
The Three Pillars
We are not building a single product. We are building a phased economic engine with three distinct revenue models, each unlocking the next:
Pillar One: Brick and Mortar (Now). This is the operating proof of concept. The cash generated here funds our R&D. This is not a loss-leading growth play. The core business is already highly profitable.
Pillar Two: The RestoreFast Exchange (Second Half 2026). This is a construction labor marketplace - a two-sided platform built on our SkillBridge and FieldMate systems that matches vetted workers to scoped projects, with a spread per side. The structural innovation is the elimination of receivables. Traditional construction operates on ~90 day payment cycles, which is the single biggest constraint on growth for every small and mid-size contractor in America. With our systems in place, our Exchange can offer same-day settlements. Zero receivables. This model turns RestoreFast from a contractor into a market maker.
Pillar Three: Enterprise Agents-as-a-Service (Early 2027). This is the licensing play. We take our validated agent runtime, the one that has been trained on thousands of real projects and real field data, and offer it as a turnkey agentic system for the 80% of the construction market (dinosaurs) that will be late adopters. Monthly or annual licensing.
Each pillar compounds on the one before it - the contracting operation generates the data, the marketplace validates the labor model, and the licensing platform scales the intelligence layer horizontally.
Why Physical Execution Is the Moat
There is a persistent misconception in Silicon Valley that software eats everything and that AI will commoditize all manual and blue-collar work into irrelevance. The opposite is happening.
AI is commoditizing software. Every SaaS tool built in 2019 is now reproducible by a competent engineer with Claude or GPT in a weekend. The barriers to entry in pure software are collapsing. Founders who raised $50 million to build dashboards are watching their products get replicated in a hackathon.
But construction? You cannot replicate a contracting operation with a prompt. You cannot build relationships with military installation commanders through an API call. You cannot deliver hazardous material abatement, mold remediation, or structural restoration through a chatbot. The physical world has friction. And that friction is the moat.
RestoreFast's defensibility comes from the combination of AI-native systems and physical-world execution. Neither alone is sufficient. A tech company with no field operations has no training data and no revenue. A contractor with no AI is trapped in the same margin compression, labor shortage, and cash cycle constraints that are suffocating the rest of the industry. The combination is what creates an asymmetric advantage.
Peter Thiel wrote in "Zero to One" that the most valuable businesses are monopolies - companies that are so differentiated that they exist in a category of one. He argued that competition is destructive and that the goal is to build something so distinctive that the question of competition becomes irrelevant. RestoreFast is the only AI-first contractor operating in production at this scale. There is no second one. No competitor has both a functioning construction business and a proprietary agent runtime trained on live field data.
We believe the period from 2025 to 2028 represents a once-in-a-generation restructuring window. Legacy enterprises are in a trough, burdened by bloated headcount, analog processes, aging workforces, and balance sheets that cannot absorb the cost of transformation. AI-first companies that are already operating in production; not pitching, not demoing, but delivering - have a three-year window to capture permanent market share before the incumbents recover.
The Metric That Matters
Every investor, operator, and analyst in the room should be watching one number: revenue per employee. Not revenue. Not headcount. The ratio. (Obviously profitability too but if you are not profitable, you are an idea, not a business.)
RestoreFast generates almost 4 times the industry average construction company revenue per employee amount.
This is the metric that Musk used to evaluate Twitter. It is the metric that Rabois predicted would define the next era of company building. It is the metric that Buffett implicitly optimizes for when he talks about asset-light businesses with wide moats and predictable cash flows.
And it is the metric that will separate the companies that survive the AI transition from the ones that get buried by it.
Construction is not exempt from the restructuring that is happening across every industry. The difference is that most industries are being disrupted from the outside by startups that have no operational footprint. Construction will be disrupted from the inside - by operating companies that embed AI into the production process itself.
That is RestoreFast. An operating company. A real contractor. With real revenue, real margins, real military contracts, and a real path to being the first billion-dollar contractor with ten employees.
The Bottom Line
The $2 trillion construction industry is facing a structural labor crisis, margin compression from material costs and tariffs, and a 90-day cash conversion cycle that strangles growth. The traditional solution - hire more project managers, buy more trucks, take on more debt - does not scale. It never has.
The AI-native solution is different: build systems that absorb the functions humans used to fill. Capture clean data at the source. Train agents on real production work. Eliminate the cash conversion cycle with a marketplace model. And then license the entire platform to the 80% of the market that will never build it themselves.
We are not waiting for the future. We are operating in it. And we are not afraid to tell you exactly how because the difficulty was never in the idea. It is in the execution. The hot, sweaty days on-site. The steel-tipped boots. The endless debate between what the project plan says and what the building actually looks like when you open the wall. The documentation. The work quality. The comeback work when something wasn't right. The change orders that blow up a budget overnight. The accidents. The unforeseen site conditions that no scope of work ever anticipated. That is the moat. Not the code.
8-figure revenue. 71% growth. 26.7% operating margin. 180% net income growth. 10 employees. 8 military bases. $30,000 in hard assets.
This is what happens when you stop brute forcing solutions, and start building systems.
This is RestoreFast :)
