First Round and Bessemer Lead Mach9’s $34M AI CAD Round

Mach9, a startup that makes AI software for civil engineers, has raised a $34 million Series A co-led by First Round Capital and Bessemer Venture Partners, TechCrunch reported on 21 October 2026. The company announced the round the same day in a blog post by its co-founder and chief executive, Alex Baikovitz, which says it brings total funding to $49 million.

In plain terms, Mach9 sells AI-powered computer-aided design, or CAD, software. It takes the 3D laser scans and images that survey crews capture of roads and sites, and produces a first draft of the engineering drawing, which licensed engineers then check and finish.

Business Pill · COST PER TASK

A short explainer on pricing AI by the job done rather than by the word. It teaches the general idea only and says nothing about any company in this story.

The Round

Prudence, an early-stage New York firm that invests in AI software for construction and real estate, also joined the round, TechCrunch reports, along with Quiet Capital, which was already an investor. Mach9’s blog post says Quiet Capital led its seed round. On our arithmetic, the company had raised about $15 million before this round, since $49 million in total minus $34 million leaves $15 million.

Neither source gives a valuation. Baikovitz told TechCrunch the money will be split between improving the product, hiring industry experts and scaling “all parts of our organization to keep up with where the market is taking us.”

From Hardware to Software

Baikovitz co-founded the company as Mach9 Robotics in 2021, with the goal of building a faster and cheaper way for civil engineers to do visual inspections, according to TechCrunch. The company built that hardware, then concluded that hardware was not the bottleneck.

“The real pain was in their workflows,” Baikovitz told TechCrunch. He dropped “Robotics” from the name and moved the company to AI CAD software. The blog post says the founders met at Carnegie Mellon’s Robotics Institute and that “two years ago, we went all-in on what they actually needed: AI CAD software for civil engineering, starting with surveying.”

Baikovitz also told TechCrunch that customers steered the change. He recalled one vice president telling him: “I’m not interested in getting this cheaper, lower-cost hardware.”

The Numbers Mach9 Gives

Mach9 says it has “nearly 100 customers,” and TechCrunch describes them as customers for the private beta of its software. The blog post names engineering firms WSP, HNTB and NV5 and state agencies including Caltrans, Minnesota DOT and the Maryland State Highway Administration.

The company says its software has been used on “more than 2,500 projects across all 50 states, surveying over 100,000 miles of infrastructure corridors.” It also says customers can “deliver surveys up to 10x faster and at a fifth of the cost.” These are Mach9’s own figures and have not been independently checked.

For the starting point, Baikovitz told TechCrunch that drafting takes “about a week per mile for someone to do manually in the legacy incumbent software that’s built by the Autodesk and Bentley Systems of the world.” One customer quote in the blog post, from Rob Ashley of the engineering firm Pennoni, says: “Mach9 enables us to do 50% more work with the same experienced team.”

Why Surveys Became the Slow Part

Mach9’s case is a labour shortage. Its blog post cites NCEES data showing that resident surveying licences fell 24% from 2011 to 2025, and it points to the $1.2 trillion that the infrastructure bill allocated to rebuilds and repairs.

A mapping vehicle can scan miles of road in a day, the company writes, but technicians still trace every curb, sign, pole and lane marking into the drawing by hand. Baikovitz told TechCrunch that Mach9 trained its own AI models to reach the centimetre-level accuracy that the work needs, because “that’s not something that a lot of off-the-shelf models are able to do or deliver at in any capacity.”

He also stressed that engineers stay in charge of the result. The software gives them tools to adjust the AI’s draft “so that they can put effectively their name behind the deliverable that they’re producing, because there is a lot of liability in this particular industry,” he told TechCrunch.

Connecting the Coverage

Our Arda piece on 20 October covered another AI startup aimed at physical engineering work: Arda, founded by former OpenAI executive Bob McGrew, raised a $160 million Series A at a $1.5 billion valuation, McGrew said on X, to help manufacturers design and run assembly lines.

Our GTC Berlin wrap this morning recorded Nvidia’s keynote examples of AI in engineering, including Dassault Systèmes using engineering agents to adapt vehicle design and agents at Siemens’ Erlangen factory testing plans in a digital twin before changes reach the factory floor.

Our Blacksmith piece today covered a $66 million round for a company that tests the code AI tools write. As we read it, Mach9 makes a similar case for engineering drawings: the AI produces the draft, and people still review it before it is used.

The Structural Read

As we read it, the round rests on the company’s own traction figures: TechCrunch and the blog post give customer, project and mileage counts, and neither gives revenue, pricing or a valuation.

As we read it, the pivot is the core of the story: Mach9 started as a hardware maker and, by Baikovitz’s account to TechCrunch, it was largely its own customers who told it they did not need better devices to capture data.

As we read it, liability shapes the product: Baikovitz told TechCrunch that engineers adjust the AI’s draft and put their name behind the deliverable, so the software is pitched as a drafting aid for licensed professionals.

Three Implications

FOR CIVIL ENGINEERING FIRMS Mach9’s blog post quotes Pennoni’s Rob Ashley saying the software lets the firm do 50% more work with the same experienced team; it is a customer quote the company chose to publish.

FOR CAD SOFTWARE INCUMBENTS Baikovitz named Autodesk and Bentley Systems as the makers of the legacy software Mach9 measures itself against, TechCrunch reports; neither company is quoted in either source.

FOR INVESTORS The $34 million round takes Mach9’s total funding to $49 million, the company says, and TechCrunch places it among startups modernising the planning-to-building pipeline, including CivilGrid, which raised $26 million in August.

The Business Engineer Lens

This story maps onto the Business Engineer framework From SaaS to AGaaS.

The framework argues that “SaaS is charged for access to tools. AGaaS charges for the execution of outcomes.”

As we read it, Mach9 sits between the two: it sells software, but the product it describes is a first draft of a survey, work a technician used to do by hand. Neither source says how Mach9 prices it, and the framework does not discuss Mach9.

What Is Not Established

Neither TechCrunch nor Mach9’s blog post gives a valuation, revenue, pricing or the size of the seed round. The customer, project and speed figures come from the company. TechCrunch calls the customers private-beta customers; the blog post does not say whether the software is generally available. We did not contact Mach9 or its investors.

The Bottom Line

First Round and Bessemer have co-led a $34 million Series A in Mach9, which says it has nearly 100 customers and more than 2,500 projects for software that drafts engineering surveys from 3D scans. As we read it, the bet is on the drawing step between scanning a road and redesigning it, where the company says skilled staff are scarce.