Mach9 pushes AI further into infrastructure mapping with Digital Surveyor 2

As infrastructure demands continue to outpace workforce growth, one San Francisco-based startup is betting that artificial intelligence can close the gap. Mach9 has unveiled Digital Surveyor 2, the latest version of its Lidar feature extraction platform – and the numbers it is putting forward are hard to ignore: project completion times up to 100 times faster, with no sacrifice in accuracy.

The release targets a well-documented pressure point in the surveying and civil engineering sector. Across telecom, transportation and utilities, the adoption of Lidar scanning technology has accelerated sharply, yet production teams have not grown at the same pace. Skilled technicians continue to spend significant portions of their time on repetitive digitization work, creating bottlenecks that limit how many projects firms can take on. Digital Surveyor 2 is designed to address that imbalance directly.

AI-powered extraction engine

At the core of the platform is an AI-powered extraction engine that suggests linear features – kerbs, pavement edges, guardrails and barriers – directly from Lidar point clouds, adapting in real time to each individual user's drawing patterns. The suggestions remain fully editable, a design choice Mach9 says is central to maintaining the quality control that engineering-grade deliverables require. The platform also includes advanced draping capabilities, improved corner snapping and integrated quality assurance tools, allowing reviewers to verify extractions in 3D without switching between applications.

Early adopters report tangible gains. Juliana Conlon, GIS engineering technician at TrueNet Communications, described work that previously took a month being completed in a matter of days. Matthew Skibba, director of reality capture at Feldman Geospatial, noted he could "shave weeks and weeks off of extraction", adding that the ability to verify vertices and lines in 3D had made his team's workflow significantly more efficient. Russell Hall, senior survey project manager at Langan, went further, calling Digital Surveyor 2 potentially the lowest-barrier-to-entry software in its category – suggesting that users with field surveying experience but limited CAD training could become productive almost immediately.

More to come

Founded in 2021 and built on research from Carnegie Mellon University's Robotics Institute, Mach9 counts Langan, Olsson, Woolpert, HDR and several state departments of transportation among its clients. CEO Alex Baikovitz described the new release not as a finished product but as a foundation, with plans to expand automation capabilities and supported feature types based on direct feedback from production teams.

For firms facing the twin pressures of rising project volumes and constrained hiring, a platform that demonstrably multiplies team output without compromising quality represents a compelling proposition – and one the broader infrastructure sector will be watching closely.