A traffic camera has a perfectly respectable reason to turn its head. An operator needs a closer look at something on the road, so the camera pans, tilts or zooms. For a person watching a screen, that movement is helpful. For software trying to measure the same intersection over time, it can make the road appear to have moved. The camera is doing its job. The software has acquired a problem.
Nicholas D'Andre encountered that problem through GridMatrix's work with the Port Authority of New York and New Jersey. Engineers were moving cameras to inspect roadway objects. His company's analytics had to accommodate the people using those cameras, and recognize when the intersection had left the picture. It developed a way to stop collecting the relevant data while the view was elsewhere and resume when the camera returned.
There is a whole business philosophy tucked into that awkward adjustment. D'Andre, GridMatrix's CEO and co-founder, wants to extract useful information from the infrastructure agencies already operate. Existing equipment arrives with existing habits, limitations and responsibilities. A camera cannot be expected to abandon its day job simply because a startup has discovered an interesting second one.
The supply chain before the street
Before he worked on transportation analytics, D'Andre worked on the arrangements that bring physical products into existence. At Apple, he was a senior global supply manager, handling sourcing strategies across suppliers of mechanical components, industrial sensing, radio-frequency test equipment and engineering services. The work connected engineering teams developing prototypes with operations teams preparing for mass production.
His responsibilities included more than half a billion dollars in annual spending across more than twenty suppliers distributed around the world. Those figures belong to his Apple responsibilities, rather than GridMatrix's revenue or his personal wealth. They give the career transition its texture: he came to infrastructure software having already dealt with the cost, coordination and practical demands of hardware.
Earlier, he had been a senior inventory manager in Amazon's retail business, using analytics to guide purchasing, inventory and distribution. During his MBA, he interned at Google and worked on demand forecasting for the first Pixel smartphone. Before those roles, he was an investment analyst in McKinsey & Company's New York office, supporting its public-equities portfolio through software, market data and financial modeling.
The settings changed; quantities kept demanding attention. What should be bought? Where should it go? What does the available data say about demand? His degrees, a BA from Pomona College and an MBA from Yale, sit alongside a career of applying analysis to decisions with physical consequences. A forecast eventually becomes a purchase order. An inventory decision eventually becomes a parcel in a warehouse.
A complaint acquires a company
The idea for GridMatrix emerged while its future founders were traveling abroad for Apple supply-chain work. They noticed responsive transportation systems and discussed improving traffic flows at home. The conversation eventually crossed the small but consequential distance between observing a problem and volunteering to take responsibility for it. GridMatrix launched in 2021, founded by former Apple engineering and operations colleagues.
D'Andre described that crossing with a compact pair of phrases: 'someone needs to do something' became 'let's do something'. The words are familiar enough to have appeared in countless conversations about traffic. The consequential part was building a company afterward. Complaining about a queue requires almost no organizational capacity. Selling a public agency something useful requires quite a lot.
GridMatrix's approach gives cameras and other sensors a further use through software. An agency's existing feeds become inputs for analytics on congestion, signal performance, emissions and safety. The company also supports sensing technologies such as radar, inductive loops and LiDAR. Its proposition is easier to understand if one starts with the installed equipment: money has already been spent gathering observations. How much more can those observations tell the operator?
“Someone needs to do something” became “let's do something”.Nicholas D'Andre
The bridge is a working examination
In April 2022, GridMatrix was selected from a global field of more than 150 startups for a Port Authority proof of concept. The route into the agency ran through the Transit Tech Lab's sustainability challenge. The initial deployment was on the George Washington Bridge. The Port Authority subsequently expanded the work to the Lincoln and Holland tunnels.
For D'Andre, those deployments put the company's premise into contact with facilities that had to keep functioning while new software was being tested. It is an appealing setting for a startup story, but the practical stakes are decidedly unromantic. Operators need information they can interpret, measurements they can trust and tools that fit the way they already work. The road keeps its own schedule.
The moving-camera problem makes that demand unusually clear. A feed is a view selected by someone for a purpose. If the view changes, software has to understand the change before its measurements become misleading. GridMatrix's response was to adapt the collection process, rather than require operators to leave their cameras untouched. The development is described in the company's public-sector proposal as an example of flexibility and risk management.
That same proposal describes D'Andre as the ultimate project manager for the company's deployments to that point. His responsibilities included project outlines, performance indicators, schedules, tasks and delivery. Client meetings were typically held every two weeks, with additional quarterly check-ins. Behind the conference appearances is a founder responsible for agendas, budgets and whether the promised work arrives on time. Infrastructure has an admirable ability to bring ambition back to the calendar.
- 01 / ObserveExisting cameras and sensors
- 02 / AnalyzeSoftware extracts traffic metrics
- 03 / ActOperators receive data and alerts
The close call becomes a record
D'Andre is also the named inventor on US patent 11,955,001, granted on April 9, 2024, for traffic near-miss collision detection. The invention addresses the problem of identifying close encounters among road users from traffic data. Cars, trucks, pedestrians and bicycles can be treated as objects whose movements and relationships are analyzed across frames.
The patent describes calculating metrics and testing conditions to determine whether a near miss or collision has occurred. One condition concerns the distance between a pair of objects. The result can be recorded and incorporated into a visualization. It is a formal treatment of an everyday event: two road users come uncomfortably close, continue onward, and leave a small piece of evidence that could otherwise disappear with the next frame.
For an operator, that distinction matters. A record of close encounters offers another way to examine a location before looking only at completed crashes. It does not make every encounter a forecast, and a patent does not establish a deployment's results. It does show a specific technical contribution by D'Andre, alongside his operational and commercial responsibilities.
The camera-movement application published in February 2025 belongs to a separate piece of GridMatrix's work. Its listed inventors are Eric Valasek and Eric Wendt. The distinction makes the team's story more interesting: a CEO can carry responsibility for adapting a project while colleagues own the invention that addresses a technical obstacle. The organization needs both kinds of work, and the names should stay attached to the right contribution.
Back to the movement of goods
In June 2024, GridMatrix announced its deployment at Port Newark and the Elizabeth-Port Authority Marine Terminal. The new assignment brought D'Andre's business back to the movement of goods, this time on the infrastructure side of the supply chain. Existing port cameras would supply a cloud platform generating alerts and data about traffic flow, vessel activity, security, emissions and other operational measures.
The port work continued the relationship that began on the George Washington Bridge. It also widened the setting. A roadway network and a marine terminal have different operating demands, even when both involve vehicles, queues and information arriving from cameras. The company described location-specific data alongside its general transportation metrics. The useful question remains what the facility's operating professionals need to know.
By May 2025, D'Andre was sharing news of a traffic-analytics pilot at Dallas Fort Worth International Airport. He credited the team's work and the airport partnership. GridMatrix's announcement described real-time video and AI being used with the aim of reducing congestion and emissions and improving safety across the airport's grounds. The pilot label matters: it describes a stage of work, rather than a completed claim of improvement.
The company also announced a platform launch at PortMiami in August 2025. Through the Miami-Dade Innovation Authority's challenge process, it was among the winners addressing cargo visibility at the port. Later, D'Andre discussed the procurement experience publicly; GridMatrix said the process had cut its time to pilot in half. Here, another kind of flow came into view: the movement of an idea through a government's buying process.
An ambition with existing equipment
D'Andre's public work extends to mobility conferences and advisory roles. His company biography lists Columbia University's Center for Smart Streetscapes and the Intelligent Transportation Society of America's Emerging Technologies Committee. His talks address cloud computing, sensor fusion and artificial intelligence in transportation. Those connections place him among researchers, public operators and technology builders dealing with the same streets from different directions.
GridMatrix's recurring ambition is to improve the safety, efficiency and environmental performance of physical infrastructure through better information. In D'Andre's career, that ambition follows years of managing supply and inventory. The founder now asks what an agency can learn from equipment already at work, and whether the answer can be delivered in a form its staff can actually use.
The traffic camera offers a pleasingly modest place to end. It has no reason to care about a company's founding story, a conference prize or a patent number. It turns when an operator needs it to turn. Making software useful around that ordinary fact is part of D'Andre's job. The second life of the camera begins with respecting the first.