Reality capture workflows have changed dramatically over the past decade.
More than a decade ago, I led the vendor team responsible for geospatial data acquisition and production for Microsoft Bing Maps before taking on a similar role supporting Uber Maps. Those were the early days of large-scale feature extraction for enterprise mapping applications and autonomous vehicles.
We were collecting imagery and lidar, building enormous production pipelines, and feeding machine learning models with countless hours of human annotation. The promise of automation was obvious, but so was the infrastructure and manual effort required to make it work.
Today, the industry faces a different challenge. Sensors have improved, reality capture has become more accessible, and organizations can collect massive volumes of lidar and imagery. The question is no longer simply how to capture the world. It is how to efficiently transform that captured data into accurate, engineering-ready deliverables.
That challenge led me to a conversation with Alex Baikovitz, co-founder and CEO of Mach9.
After discussing surveying workflows, CAD, cloud computing, and production environments, one thing became clear: the larger story is not simply another advancement in artificial intelligence. It is the evolution of geospatial production itself.
Reality Capture Has Outgrown Traditional CAD
“CAD was designed to manually draw and draft information. It was never designed for automation. It was never designed to QAQC or validate the correctness of automated results.”
That observation from Baikovitz highlights a fundamental shift taking place across surveying and engineering.
Traditional CAD systems were built around the assumption that humans create geometry. Reality capture has changed that process. Projects increasingly begin with billions of lidar points, high-resolution imagery, mobile mapping trajectories, SLAM scans, drone collections, and aerial surveys.
The challenge is no longer creating a digital representation of the world from scratch. It is interpreting, validating, refining, and delivering the world that has already been captured.
Mach9’s Digital Surveyor reflects this shift by placing reality capture data at the center of the production environment. Rather than treating point clouds and imagery as files imported into a drafting system, the platform is designed around extracting information, editing results, performing QA/QC, and producing engineering-ready deliverables.
The Bottleneck Has Moved From Collection to Production
During the early days of autonomous mapping, collecting enough high-quality data represented one of the largest challenges. Sensors, storage, and computing resources were expensive, and organizations needed significant infrastructure to manage large-scale mapping programs.
Those constraints have changed.
Mobile mapping systems, drone platforms, static scanners, SLAM systems, and aerial lidar have matured significantly. The industry has become highly capable at collecting reality, moving the bottleneck downstream.
Engineering firms are increasingly challenged by how to transform growing volumes of reality capture data into consistent deliverables that meet professional standards for accuracy, quality, and turnaround time.
Baikovitz describes Mach9 not simply as an AI company, but as a production platform. Automated feature extraction is one component of a larger workflow that includes editing, validation, templates, collaboration, engineering standards, and quality assurance.
Cloud-Based Production Changes the Workflow
The shift is not only about automation but optimizing the underlying production architecture.
Historically, large-scale geospatial production required significant internal infrastructure, including dedicated workstations, storage systems, databases, specialized software environments, and engineering resources.
Mach9 approaches production differently through a cloud-based environment accessed through a browser. This model allows organizations to scale computing resources as needed, collaborate within a shared production environment, maintain consistent software versions, and reduce the operational burden of maintaining complex infrastructure.
The change reflects a broader transformation across software industries. Instead of every organization building its own production environment, specialized platforms can provide the infrastructure needed to support sophisticated workflows.
Automation Still Requires Human Expertise
While automation is becoming increasingly important in geospatial workflows, surveyors and engineers remain responsible for the accuracy and reliability of their deliverables. That makes verification and quality control essential.
“Faster extraction is not enough. Surveyors need to verify the result and QAQC efficiently.”
Mach9’s Trajectory View addresses this challenge by connecting calibrated imagery directly with lidar point clouds. Users can inspect extracted features, compare imagery with point cloud data, and validate results without leaving the production environment.
The goal is not to replace professional judgment, but to allow experts to spend less time performing repetitive tasks and more time applying the experience required to deliver trusted information.
As Baikovitz explained, “Mach9 does not replace surveyors, we empower them.”
Production Becomes Geospatial Infrastructure
The most significant shift may not be AI or any individual software feature. It may be the recognition that production itself has become a critical layer of geospatial infrastructure.
Historically, organizations building large mapping platforms or working with large data sets had to invest heavily in internal production systems. They developed custom pipelines, managed storage and computing resources, built quality assurance processes, and maintained specialized teams.
Today, platforms like Mach9 represent a move toward making those capabilities available more broadly. Surveying firms, engineering organizations, utilities, and infrastructure owners can increasingly access sophisticated production workflows without building the entire technology stack themselves.
The advantage shifts from maintaining complex systems to applying expertise, solving client problems, and delivering better outcomes.
Solving Practical Problems
One of the most consistent themes throughout the conversation was a focus on practical workflows.
Rather than technology for its own sake, Mach9 is focused on helping surveyors, engineers, and infrastructure professionals complete projects more efficiently while maintaining confidence in their deliverables.
The future of reality capture production depends on a combination of automation, cloud infrastructure, and human expertise. Organizations will continue collecting more detailed information about the physical world. The challenge will be transforming that information into accurate, trusted, and usable data.
Mach9 represents one example of how the geospatial industry is responding to that challenge.
Another Recent Article: Reality Capture Game Teaches Lidar Scanning Trade-offs
