The company teaching software to read a bridge the way an engineer wishes they could - precisely, and with a memory.
Caption There's no glamour shot here - just a wordmark and a readout, which is fitting. Aren spends its days looking at things most of us drive over without a thought: the underside of a bridge, the seam in a tunnel, the hairline crack that was two millimeters last spring and is three now. Point a camera. Wait. The verdict comes back as a number.
Here is a fact that sounds made up and isn't: somewhere in the world, a bridge fails roughly every three days. In the United States, more than 56,000 bridges are rated structurally deficient, and something like 200 million people cross an at-risk span every single day on their way to work, school, or the grocery store. We have collectively decided this is fine, in the way you decide a check-engine light is fine as long as the car still starts.
Aren, a New York company founded by the structural engineer Ali Khaloo, is built around a quieter observation. The problem with aging infrastructure is not entirely that we lack money to fix it - though the unfunded gap runs to about $2 trillion in the US and roughly $15 trillion globally. The problem is that we don't reliably know which things to fix first. Inspection, for about a century, has meant an engineer, a truck, a clipboard, and a judgment call. Two inspectors can look at the same beam and grade it differently. Next year, a third inspector shows up with no memory of what the first two saw.
Aren's pitch is that this is fundamentally a data problem, and data problems have data solutions. Its platform is source-agnostic: feed it smartphone photos, drone video, laser scans, infrared, or sensor feeds, and it stitches them into a high-resolution 3D digital twin of the asset. Then computer-vision and deep-learning models do the part humans are bad at - detecting damage, measuring it in consistent units, and comparing today's crack to last year's crack. Deterioration stops being an opinion and becomes a trend line.
That last part is the trick worth sitting with. Aren's real product isn't a camera or a drone; the industry already drowns in imagery. What it sells is interpretation and memory - the layer that turns a pile of raw pixels into a decision an asset owner can defend, and a record that persists so that next year's inspection is a comparison rather than a fresh guess.
The commercial logic follows from there. If you can measure condition consistently and forecast where it's heading, you can also generate the things infrastructure owners actually need to act: maintenance schedules, capital-allocation plans, and service-life cost estimates. Roughly $9 trillion a year gets spent on infrastructure worldwide, and a large slice of it lands in the wrong place. Aren is, in a sense, a company built to aim that money better.
The pipeline is deliberately unglamorous. It starts with whatever data you already have and ends with a spreadsheet a budget committee can act on.
Photos, drone video, laser scans, infrared, and surface or sub-surface sensors - no specialized rig required.
Raw data is stitched into a high-resolution virtual replica of the asset, available on demand.
Computer vision finds damage and measures it in consistent units, then tracks change over time.
Predictive models forecast deterioration and output maintenance schedules and capital plans.
Aren describes the scale of the problem in figures that are hard to picture. Here they are, side by side. Bars are scaled for illustration.
Figures as stated publicly by Aren; bar lengths are indicative, not to precise scale.
A computer-vision and machine-learning toolset that detects and quantifies damage on bridges, tunnels, roadways, and other civil assets - cost-effective and repeatable where manual inspection is slow and subjective.
High-resolution virtual models of physical assets, available on demand, so an owner can inspect a structure remotely and keep a permanent visual record of its condition.
Deep-learning models that track deterioration over time and forecast future structural change - turning inspection from a snapshot into a moving picture.
Maintenance schedules, capital-allocation recommendations, and service-life cost estimates generated from the underlying assessment data.
A structural engineer by training, Khaloo earned his PhD in Structural Engineering at George Mason University and a master's at Tufts. He spent more than a decade at the intersection of robotics, computer vision, autonomous sensing, and AI - the specific and unusual mix a company like Aren requires. He has authored 20+ peer-reviewed papers and holds patents on the automated generation and AI-powered analysis of 3D models of large civil infrastructure. Aren was incubated through Cornell Tech's Runway startup program on Roosevelt Island, a short walk from where its data pipelines now watch structures age.
Incubated through the Runway Startups postdoc program on Roosevelt Island, translating a decade of structural-engineering research into a product.
Closed a $2.4M seed round; total funding to date reaches roughly $4.77M.
Won the toll-road operator's inaugural challenge for AI-powered drone inspection and automated damage quantification.
Won Demo Day at the Infrastructure Conference for its AI digital-twin and asset-intelligence platform.
Continues to promote its civil-infrastructure management platform to asset owners as bridge-safety incidents keep the topic in the news.
Aren's office sits at 2 West Loop Road - on Cornell Tech's Roosevelt Island campus, in the middle of the East River.
The platform is source-agnostic: it can start from an ordinary smartphone photo, no specialized hardware needed.
Its founder has published 20+ papers and holds patents on AI analysis of 3D infrastructure models.
Aren frames its edge as memory, not cameras - the twin remembers exactly how a crack looked a year ago.
Profile compiled from public sources. Figures approximate where noted. Aren is an independent company; this page is an editorial profile.
Aren is a New York-based AI company that helps owners of bridges, tunnels, and other large civil infrastructure figure out what is quietly falling apart before it becomes a headline. Its platform ingests source-agnostic data - smartphone photos, drone video, laser scans, infrared, and sensors - and turns it into 3D digital twins that detect and quantify damage, track deterioration over time, and produce maintenance and capital-allocation plans. Founded by structural engineer Ali Khaloo out of Cornell Tech, Aren pairs computer vision and deep learning with decades of structural mechanics to give asset owners a quantitative, repeatable way to decide where to spend limited repair dollars.
Last updated: