It’s Ingenious: Can Engineering Solve the Housing Crisis?

There’s so little we agree on as a global society these days. Whether the divide is political, geographical, or cultural can vary, but that constant is that there is no constant. There are almost no meaningful issues where we can find nearly uniform alignment. Oh, wait, here’s one:

We need more housing. And we need a lot of it.

According to a National Association of Realtors report from 2021, the U.S. needs between 5.5 and 6.8 million new housing units to meet current demand, and that number grows by approximately 1.7 million units annually, the National Association of Home Builders estimates. In 2025, less than 1.5 million units were completed, which means we’re not keeping up with demand, let alone making progress on the backlog of units we require. Given this supply/demand imbalance, it’s not surprising that home prices in the U.S. have increased sharply. Peak growth, as most of you reading this are undoubtedly aware, came during the pandemic, when the perfect storm of stay-at-home work and life paradigms, cheap debt, and government stimulus led to home prices escalating by 41% during a 30-month period from 2020 to 2022.

And the news is the same all over. In the UK, the Office for National Statistics estimated that home prices rose 90% between 2010 and 2022; in Australia, the Bureau of Statistics said the Residential Property Price Index reflected an increase of approximately 130% between 2003 and 2022.

You get the idea. And there’s no need to belabor these points further, because, as noted at the outset, this is one area where there is almost universal agreement. The query isn’t: “is there a problem?” It’s “What do we do about it?”

There’s no single answer to that question, and any solutions will undoubtedly require contributions from politicians, developers, and many other constituencies. But when pondering a remedy, it’s unlikely that many of you found your thoughts gravitating towards the critical role engineers play in this process. And, until a recent conversation I had with a gentleman named Steven Cassells, my contemplations wouldn’t have trended in that direction either.

But that’s what “It’s Ingenious” is all about—new solutions to old problems.

Cassells is an engineer, and he’s also a Director at Neuron, a technology company that’s built a platform enabling the production of engineering reports at an accelerated pace, without sacrificing accuracy in the process. Neuron debuted in Sydney, Australia, expanded to London two years ago, and this year opened a location in New York.

Cassells told me that particular map of expansion was always the plan. By covering those three markets, the platform encompasses both the metric and imperial measurement systems, as well as the two major global bodies that govern engineering standards, ASHRAE and CIBSE.

The technology that Neuron has developed can be used for a wide array of applications, but the capability that Cassells is most passionate about is its capacity to alleviate the housing crisis. That potential has already been partially realized in the Australian market, where Cassells said Neuron’s platform has been used on approximately 600 projects totaling about $80 billion worth of primarily mixed-use and multifamily buildings; about 25% of that work has been for affordable housing.

While the geographic growth of Neuron’s business has been intentional, its expanding scope has largely been organic. Cassells said he and his colleagues were initially interested in merely automating or accelerating some of the more repetitive tasks associated with their roles. Their ambition has evolved a bit.

“What’s happened since we launched the business is we’ve ended up in this entirely new place that we didn’t believe we would end up in, which is this world of design optimization,” Cassells said. “It’s been organic, and it’s been largely driven by a terrible, terrible global cost crisis where the cost to deliver a home is probably twice what it was five, six years ago. But because we’re all constrained by mortgages and the cost of living generally has gone up, we can’t afford to pay double for our home.”

I talked to Cassells about how we can get smarter about building properties and, in the process, make homes more affordable. So, today, we’ll be building with blocks that include:

• Developments don’t go wrong, but start wrong

• The most impactful fixes aren’t sexy

• Unintended consequences of good intentions

• The developer is the good guy (How many of you are relieved to read that sentence? And in a piece about affordable housing no less!)

Cassells and his colleagues at Neuron have a kind of motto they like to use about what they view as the most fundamental issue impacting property development.

“Projects don’t go wrong,” he told me. “They start wrong.”

He added that, sometimes, the best advice he can give as an engineer, or that someone can garner from Neuron’s platform is “just don’t buy that site.”

“Don’t build over there. It’s contaminated land that’s got floodplain issues. There’s no utilities.”

Unfortunately, Cassells explained that all too often developers don’t heed that kind of advice. He said he saw it in Sydney all the time. A developer decides to build a massive property four clicks from the nearest sewer, for instance, which means that now a truck has to go out there once a week to empty waste tanks, and that will continue to be the case for four years, maybe even longer, until someone can get the sewer hooked up. And that cost, of the truck, and of the sewer hook up, all of that’s getting passed, in one way or another, to the person buying a unit in that building.

But let’s say the site isn’t contaminated or lacking utilities. Even then, Cassells said that many developers “design from the inside.” They get bedazzled by an architect’s wondrous sketches, and then they force the engineering aspects of the development to conform to those designs, as opposed to the other way around. This approach almost always results in added costs and increased time.

And in a high-interest environment those two things, as is so often the case, are one and the same. Every month of additional time that it takes to design or build a property is an additional month of interest that, once again, is typically borne by the end user.

Cassells noted that this scenario is frequently compounded by well-meaning city and local officials. No one wants another drab, boring building in their neighborhood. They want something that captures attention, that makes people proud to be denizens of that municipality. The problem, though, is that such architectural flourishes create practical costs, and in the present moment those can be hard to justify.

Of course, Cassells acknowledged that all this depends on the market you’re building in and the customer you’re building for. And this is where things get particularly problematic for developers focused on affordable housing.

“There’s three parts to property development,” Cassells said. “Buy a piece of land, build a building, and sell it for 20% more than what it costs to buy the land and to build a building. So, in Manhattan, you can sell the building to billionaires and millionaires, which means you can sell it for so much money that it easily makes your 20% margin even if the building’s expensive and even if the land is expensive. That margin goes down as you go further and further away from desirable, high wealth areas, which means it gets harder and harder. And at the very bottom of that spectrum is social affordable housing, the type of housing that makes the least financial sense because it’s still expensive to build a building, but you’re selling or renting it for the lowest possible margin.”

Cassells’ solution to this problem isn’t necessarily radical. In fact, it might seem relatively obvious, but its conspicuousness doesn’t necessarily mean it’s being implemented as often as you might expect.

“Start with the feasibility first, and then design to that feasibility.”

So, how do you enhance feasibility? If you were expecting answers that talked about AI or radical building materials (did you know asparagus can actually be used in place of concrete?) those aren’t the solutions Cassells is most interested in.

Neuron’s platform was built before LLMs rose to prominence, and the company still doesn’t use AI to conduct any aspect of its engineering assessments for one very simple reason: the tech just isn’t up to the task yet.

“We really like AI,” Cassells said. “We like the conversational nature of taking data and coming back with something bespoke to a client’s specific question. But there’s no way it’s reliable enough to design structures for building and engineering systems today. So, we like to put AI over Neuron, not under Neuron.” In other words, AI is leveraged for the platform’s user interface, but nothing else.

So, if this is one area where AI can’t save us—at least not yet—what are some of the fixes that Cassells has seen work effectively?

The most impactful, as well as obvious, one is building a property that maximizes the amount of space that can be used for units, thereby increasing the developer’s yield. It may seem counterintuitive that a developer would ever take another approach, but it goes back to the issue of the design leading the process, as opposed to the more practical elements coming first. Cassells said they often consider something he referred to as the facade-to- floor-space ratio. He explained that tall, skinny buildings, particularly those with units on just one side of a corridor, are “terrible” from a yield perspective.

“You build a whole building’s facade, which costs you a lot of money, but you only have half a building’s worth of apartment.” Building a more square-shaped property, with a corridor that can serve apartments on either side, is a more practical, albeit arguably less elegant, solution.

Another fix is building more sustainable properties. A building that is engineered to use less energy costs its end user less money, which means they’re willing to pay more to buy or lease the unit, because their all-in expenses are the same.

In more affordable properties, Cassells also advocated for making features such as outdoor space and amenities like laundry rooms part of common areas instead of housing them within individual units. This approach has the dual benefit of reducing costs for the developer, while also fostering a greater sense of community within the property. People are more likely to get to know their neighbors if they’re sharing a laundry room, roof deck, or garden, than if everyone is enjoying their own balcony and patronizing individual laundry facilities.

Perhaps the most important way that developers can make projects more feasible though, is by iterating different designs and scenarios until they find the one that allows them to build the most units with the greatest efficiency. And that’s where Neuron’s platform comes in. By making the engineering process faster and more accessible, Neuron enables developers to iterate designs in a manner that is both fast and affordable.

“We think that the last part of this puzzle is for everyone to be able to move faster and iterate faster until they find that variation of the design that maximizes their yield.”

Without a platform like Neuron, that’s just not a feasible approach (pun intended) for most developers. It takes too much money and too much time (which, as discussed earlier, also means more money).

Cassells also mentioned the viability of pre-fabricated housing units and construction components, which he said Neuron had worked with fairly extensively, but he agreed that one of the biggest issues in developing affordable housing isn’t the materials, but the political and cultural ecosystem in which the building resides.

So, full disclosure: I come from a developer family. My father and my uncles are developers, as was my grandfather. They created a business that I still derive economic benefit from, even if I’m not actively involved in it. In other words, I’m predisposed to be sympathetic towards their ilk.

Here’s the thing though: developers aren’t the enemy when it comes to making housing more affordable. In fact, as Cassells noted, they’re kind of the heroes of this story.

Imagine, he said, if developers just took their money out of development and put it into the stock market. They could do that and, if history is any indication, enjoy a comfortable 10% yield. But the rest of society would be poorer for it. We would lose offices, and retail buildings, warehouses, and, of course, houses. In order to make affordable housing viable, we need to accept that developers, even the not-for-profit organizations that often construct social housing, need to be able to generate a 20% return. If that doesn’t happen, a project doesn’t go forward.

“A bank will never fund the project unless it’s viable,” Cassells said. “So, a lot of people think, ‘oh, the greedy developers, they’ll just have to suck it up.’ That’s not how it works. They just don’t do the project.”

Similarly, and oh how it wounds my left-leaning heart to admit this, regulations and building codes need to be implemented more thoughtfully and stripped back in some instances. Cassells said that local municipalities should be more thoughtful about how new or revised codes will impact feasibility. He said that, recently, a developer told him that “If you have an affordable housing issue, whenever there’s a code change proposed, irrespective of its merits, the only question that should be asked is, will this make it more expensive to deliver housing?” While Cassells acknowledged that perspective was a bit blunt, he found it hard to argue with.

Cassells and I agreed that the situation is improving when it comes to how politicians and government officials regard development—nimby is giving way to yimby, thank your deity of choice. But more education is required to really help politicians and average citizens alike understand how development works and what the barriers are that actually constrain the housing supply.

For example, Cassells referenced the so-called “99-rule” in New York City, which requires developers that are constructing a building totaling 100 units or more to employ labor from unions that demand a higher rate. In theory, this seems like a potentially good regulation that encourages larger developers to pay higher wages. In practice, it means that most New York City developers stop at 99 apartments, which decreases the housing supply and drives costs up. “That’s a completely arbitrary rule that, if it was removed, you would immediately have more density,” Cassells said.

This disconnect, between intentions and outcomes, is part of what Cassells hopes Neuron can help ameliorate. By putting engineering at the forefront of the design and development process, and by making it easier and cheaper for developers to iterate designs until they find one that actually pencils, we can make housing more affordable for everyone—even developers themselves.