Nominal

Nominal is a software company for people who build things that can kill you if they fail. Its customers test autonomous vehicles, rockets, and energy systems—hardware where a missed anomaly isn't a bug fix, it's a recall, a crash, or a failed mission. The company's pitch is that the bottleneck in bringing these systems to market isn't the hardware itself, but the messy, manual, often siloed process of testing it. And with a recent $80M raise at a $1B valuation, the market is signaling that this thesis—that hardware testing is fundamentally a software problem—is worth taking seriously.

The $1B bet that hardware testing is a software problem

For decades, hardware engineering has lagged behind software in tooling. Software teams have version control, continuous integration, and observability platforms; hardware teams have Excel spreadsheets, ad hoc scripts, and tribal knowledge. Nominal's founders, led by Cameron McCord, are betting that the same kind of data infrastructure that transformed software development can do the same for hardware. The company's tagline—"Software for engineers to build, test and validate complex hardware systems"—is deliberately broad, but the focus is clear: the test phase, where hardware proves it works or doesn't.

The problem is acute. A single test run on a vehicle or a rocket generates terabytes of telemetry, video, and logs. Analyzing that data is slow and fragmented: one engineer might be looking at a CSV export, another at a video replay, and a third at a PDF report. By the time the team agrees on what happened, days have passed. Nominal's answer is to bring all that data into one place, with real-time analysis and collaboration tools, so that the gap between "test happened" and "we know what it means" shrinks from days to minutes.

The two products that make up Nominal's stack

Nominal is not a single tool but a connected suite, and understanding the distinction between its two main products is key to seeing how it fits into a hardware team's workflow.

Nominal Core is the cloud-based data platform. It ingests telemetry, logs, video, and simulation results into a searchable, structured catalog. Engineers can monitor live data streams, set alarms, and track system health in real time. After a test, they can replay synchronized data, analyze trends over time, and generate reports. Core is the "single source of truth" for engineering data across the lifecycle, from prototype to production.

Nominal Connect is the edge compute platform. It runs where the hardware lives—on a lab bench, a test range, or a factory floor. Connect lets engineers deploy test stands in minutes, not months, and write Python-based test apps that control instruments and sequence tests. It handles the real-time, deterministic execution that can't be done in the cloud, and it can operate fully offline, syncing data to Core when connectivity returns.

The two products are designed to work together: Connect executes the test, Core analyzes the results. But they can also be used independently, which gives Nominal flexibility in how it enters an organization.

What Connect does that test scripts cannot

Anyone who has worked in hardware testing knows the pain of writing a test script that only one person can run. The script lives on a laptop, depends on a specific instrument driver, and produces a text file that no one else can interpret. Connect is an attempt to industrialize that process. It turns Python scripts into "operator-ready applications" with live interfaces, state management, and hardware orchestration. The promise is that an engineer can write the test logic, and Connect handles the rest—generating UIs, managing state, and sequencing hardware.

This is a significant shift. Instead of a test being a one-off artifact, it becomes a repeatable, deployable application. The same test code can run across dozens of manufacturing stations without modification, ensuring consistency from prototype validation to production line testing. For teams that need to prove reliability at scale—say, a defense contractor producing hundreds of autonomous vehicles—this consistency is critical.

Connect also addresses a practical constraint: many test environments are air-gapped or have unreliable connectivity. By running locally with deterministic timing, Connect ensures that tests aren't dependent on a cloud connection. This is a feature that resonates in defense and energy, where security and reliability are paramount.

The workflow Nominal actually sits inside

The hardware lifecycle is long and iterative: design, prototype, test, refine, produce, operate. Nominal's software is designed to sit inside the test and validation phases, but with an eye toward the entire lifecycle. The website describes a flow: Test, Manage, Monitor, Analyze, Automate. This is not just a feature list; it's a workflow that mirrors how hardware teams actually work.

During a test, engineers use Connect to execute and monitor in real time. The data streams to Core, where the team can see live telemetry, alarms, and system health. After the test, they replay synchronized data, analyze trends, and generate reports. Over time, Core builds a catalog of engineering data that can be used to track configurations, software versions, and field performance. This "trend over time" capability is crucial for spotting issues early and extending reliability.

Nominal's value proposition is that it collapses the time between test and insight. The Anduril case study on the website claims that review time dropped from over five hours to near real-time across 300+ engineers. While that's a single data point, it illustrates the core promise: faster feedback loops mean faster iteration, and faster iteration means faster progress.

The name 'Nominal' and the precision it promises

The name "Nominal" is a clever piece of branding. In engineering, "nominal" means "within expected parameters"—a system is operating nominally when everything is as it should be. The name signals precision, correctness, and control, which are exactly the values Nominal wants to associate with its software. It's a word that engineers already use, so it feels native to the domain.

The domain, nominal.io, is short and memorable, and the brand is consistent across the website, with a clean, technical aesthetic. The name also has a subtle double meaning: "nominal" can also mean "in name only," which could be a risk if the product fails to deliver. But for now, the engineering connotation dominates, and it positions Nominal as the tool that helps teams achieve "nominal" status.

The risks of scaling from Anduril to everyone else

Nominal's customer list is impressive—Anduril, Regent, Hermeus, Scout AI—but it's also narrow. These are cutting-edge companies in defense and aerospace, where the willingness to adopt new tools is high and the cost of failure is extreme. The harder question is whether Nominal can expand beyond this early adopter base to more traditional manufacturers in automotive, energy, and industrial robotics.

Those industries have legacy systems, conservative engineering cultures, and existing test infrastructure. Nominal's open architecture and integrations with tools like AWS and Azure are a step in the right direction, but the sales cycle is likely long. The company's move to open a Seattle office, as reported by GeekWire, suggests it's serious about scaling, but it also faces competition from established players in the test and measurement space, as well as from internal IT teams that might prefer to build their own solutions.

Another risk is the breadth of the platform. Nominal is trying to be both a data platform and an edge compute platform, which is a lot to execute on. The company's success will depend on whether it can deliver on both fronts without spreading itself too thin. The $80M raise gives it runway, but the hardware testing market is notoriously fragmented, and winning it will require more than just good software—it will require trust.

Still, the thesis is compelling. If hardware testing is indeed a software problem, then Nominal is well-positioned to be the standard platform for solving it. The name promises precision, and the product aims to deliver it. For now, the bet is that engineers will trade their spreadsheets for a tool that helps them achieve "nominal" status—and that's a bet worth watching.