Mach9
AI CAD software for civil engineering, automating the conversion of 3D data into structured maps.
Website: https://mach9.ai/
Cover Block
Open sources
| Field | Value |
|---|---|
| Name | Mach9 |
| Tagline | AI CAD software for civil engineering, automating the conversion of 3D data into structured maps. |
| Headquarters | San Francisco, United States [Y Combinator] |
| Founded | 2021 [Y Combinator] |
| Stage | Series A [Morningstar / PR Newswire, October 2026] |
| Business model | SaaS |
| Industry | Deeptech |
| Technology | AI / Machine Learning |
| Geography | North America |
| Growth profile | Venture Scale |
| Founding team | Co-Founders (4): Alex Baikovitz, Haowen Shi, Michael Mong, Zachary Sussman [Morningstar / PR Newswire, October 2026] |
| Funding label | $50M+ |
| Total disclosed | $49,000,000 [Morningstar / PR Newswire, October 2026] |
Links
Open sources
- Website: https://mach9.ai/
- Y Combinator: https://www.ycombinator.com/companies/mach9
- YouTube: https://www.youtube.com/watch?v=O0u27FY1sCc
- Apple Podcasts: https://podcasts.apple.com/us/podcast/lidar-magazine-podcast/id1728371618
What an Investor Needs First
PUBLIC Mach9 builds AI CAD software for civil engineering, turning large-scale 3D and geospatial survey data into structured maps and engineering deliverables, and it merits attention now because the company has paired that thesis with a $34 million Series A announced in October 2026 and reported early usage across nearly 100 private beta customers and more than 2,500 projects [TechCrunch, October 2026] [Morningstar / PR Newswire, October 2026]. Founded in 2021, the company emerged from Carnegie Mellon Robotics Institute ties and appears to have pivoted from an earlier reality-capture and robotics orientation toward a narrower software wedge in geospatial production, a shift that matters because it places the business closer to recurring workflow software rather than hardware-enabled services [TechCrunch, October 2026] [The Geoholics podcast, September 2025] [Y Combinator].
The product case is straightforward: Mach9's Digital Surveyor software is positioned to automate the labor-intensive step between raw scans and usable CAD or GIS output, with the newer Digital Surveyor 2 described as a web-based CAD application for surveyors, engineers, and drafters working on engineering-grade mapping projects [Informed Infrastructure, March 2026] [Mach9] [Forbes, January 2026]. The differentiation claim rests less on a novel model layer than on embedding AI inside civil-engineering and survey workflows where accuracy, operator control, and infrastructure-scale throughput determine whether automation is actually adopted [Geo Week News] [LIDAR Magazine Podcast] [TechCrunch, October 2026].
The founding bench is relevant to that problem set. Public materials name Alex Baikovitz, Haowen Shi, Michael Mong, and Zachary Sussman as founders, and the Series A announcement says the group previously worked on perception and navigation technology at Tesla, SpaceX, Apple, NASA, and Google X, which is credible preparation for machine perception problems but not, by itself, proof of enterprise distribution strength [Morningstar / PR Newswire, October 2026] [Crunchbase] [Highperformr].
On capitalization, the cleanest public read is that Mach9 has disclosed $49 million in total funding after its October 2026 Series A, with Quiet Capital leading the earlier $12 million seed and First Round Capital and Bessemer Venture Partners co-leading the Series A; the business model is described as SaaS, and Y Combinator is also listed among backers [Morningstar / PR Newswire, October 2026] [Mach9] [Y Combinator]. One caution is that publicly reported funding totals do not reconcile perfectly across sources, with Forbes listing $15 million raised as of January 2026, so investors should treat chronology carefully rather than assuming every intermediate financing event is fully visible [Forbes, January 2026] [Morningstar / PR Newswire, October 2026].
Over the next 12 to 18 months, the key watchpoints are whether private-beta usage converts into durable production deployments, whether named engineering and transportation customers translate into broad repeatability across agencies and design firms, and whether the company can substantiate automation claims with evidence that holds up in high-consequence engineering workflows [TechCrunch, October 2026] [Morningstar / PR Newswire, October 2026] [Mach9]. The current public record supports a company with a real technical wedge and meaningful investor backing, but the adoption question is still about operational proof at scale rather than concept validation.
Partially corroborated -- Relies on a mix of independent reporting, company materials, and partial corroboration across TechCrunch, Morningstar / PR Newswire, Forbes, Y Combinator, and company sources.
Taxonomy Snapshot
| Axis | Value |
|---|---|
| Stage | Series A |
| Business Model | SaaS |
| Industry / Vertical | Deeptech |
| Technology Type | AI / Machine Learning |
| Geography | North America |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (3+) |
| Funding | $50M+ total disclosed (approximately $49M) |
Inside the Company
PUBLIC
Mach9 presents as a San Francisco-based infrastructure software company founded in 2021, with its public positioning now centered on AI CAD software for civil engineering rather than the broader robotics language seen in some early descriptions [Mach9] [Crunchbase]. The public record is thin on legal-entity detail, but the company website and Crunchbase align on the core operating profile: a venture-backed software business building tools that convert geospatial and survey data into engineering workflows [Mach9] [Crunchbase].
The clearest chronology begins with the 2021 founding and the team’s later public rollout around Digital Surveyor, the company’s mapping platform [Crunchbase] [Mach9]. By March 2026, Mach9 was publicly introducing Digital Surveyor 2 as a web-based CAD application, which marks a more explicit product framing around end-to-end engineering-grade mapping inside a single environment [Mach9]. By October 2026, the company was publicly describing a $34 million Series A that brought total disclosed funding to $49 million, a milestone that also signals a sharpened go-to-market focus around civil engineering software rather than its earlier reality-capture orientation [Mach9] [Crunchbase].
Verified against public records -- Confirmed by Crunchbase and the company website.
Under the Hood
MIXED
Mach9’s product story is straightforward on the surface and more consequential once the workflow is unpacked. The company describes itself as building AI CAD software for civil engineering, with software that converts large-scale 3D and geospatial data into structured maps and CAD workflows rather than leaving survey teams to draft from raw scans by hand [Mach9] [Forbes, January 2026]. Public product descriptions consistently center on the same task: turning reality-capture inputs into engineering-ready outputs for surveyors, engineers, and infrastructure owners, even if the exact phrasing varies across company, press, and profile sources [Forbes, January 2026] [Y Combinator] [Crunchbase].
The named product is Digital Surveyor, which Mach9 says automates creation of CAD and GIS maps from mobile mapping system data and can identify infrastructure elements such as utility poles, road signs, and traffic lights [Mach9]. In March 2026, the company introduced Digital Surveyor 2 as a web-based CAD application intended to let surveyors, engineers, and drafters complete engineering-grade mapping work in a single environment [Informed Infrastructure, March 2026]. Some performance claims are more directional than independently verified, including company statements that customers have surveyed more than 100,000 miles and that extraction or drafting can be materially faster than manual workflows, so those figures are better read as vendor claims than settled benchmarks [Mach9].
The technical posture appears to emphasize applied perception and production usability rather than a generic model layer. Founder and company commentary points to geospatial workflow automation, with attention to trust, control, and scale in how AI-assisted outputs are reviewed and used in practice [Geo Week News] [Apple Podcasts]. On the implementation side, one surfaced software engineering role referenced sensor-data integration, geospatial datasets, and Digital Surveyor, which supports a reasonable view that the stack is built around processing survey and mapping data pipelines rather than lightweight document automation (inferred from job postings) [Mach9].
Partially corroborated -- Product scope is corroborated by company materials, Forbes, Y Combinator, and Informed Infrastructure, but several capability and performance details remain company-described rather than independently benchmarked.
Market Research
PUBLIC
The market matters now because civil engineering and infrastructure owners are under pressure to turn growing volumes of reality-capture data into design-ready outputs faster, while labor-intensive drafting and mapping workflows remain a bottleneck [TechCrunch, October 2026] [Forbes, January 2026].
Mach9 is positioned at the intersection of several established budgets rather than inside a neatly disclosed standalone category. The public record supports demand from surveyors, engineers, infrastructure owners, and transportation agencies that need raw 3D or geospatial data converted into maps, CAD files, and engineering inputs [Forbes, January 2026] [TechCrunch, October 2026]. What is less clear, and worth stating plainly, is that no named third-party market report in the cited material provides a direct TAM, SAM, or SOM for AI CAD software in civil engineering, so any market sizing here has to be read through adjacent categories rather than as a verified category total.
That adjacency still says something useful. Mach9's product description places it between geospatial production software, civil CAD workflows, and infrastructure asset mapping, with a stated focus on turning point clouds and mobile mapping data into engineering deliverables [Informed Infrastructure, March 2026] [LinkedIn, March 2023]. Its customer references also span private engineering firms such as WSP, HNTB, and NV5, and public agencies including Caltrans and Minnesota DOT, which suggests the spend authority may come from a mix of design software, surveying, transportation engineering, and asset management budgets rather than a single line item [Morningstar / PR Newswire, October 2026] [Citybiz, October 2026].
The clearest tailwind in the source set is operational, not thematic. TechCrunch reported that Mach9 had nearly 100 private beta customers and had been used on more than 2,500 projects by October 2026, which indicates recurring demand for workflow compression in production mapping rather than a purely experimental AI deployment [TechCrunch, October 2026]. Company materials sharpen the same point, though with lower evidentiary weight, claiming customers have surveyed more than 100,000 miles and that the software reduces extraction and drafting time materially [Mach9] [Mach9, retrieved 2026]. Those company claims should be treated as directional until independently corroborated, but they align with the broader market logic that firms handling infrastructure mapping are trying to automate scarce skilled labor.
A second tailwind is public infrastructure spending and maintenance complexity. Mach9's own framing, echoed in a Quiet Capital essay by CEO Alexander Baikovitz, is that better and more timely geospatial data can improve infrastructure modernization economics [Quiet Capital]. That does not prove budget conversion on its own, but the named customer mix, including state transportation agencies and large engineering consultancies, fits a market where road, utility, and public works programs generate sustained demand for surveying, drafting, and asset documentation across long project cycles [Morningstar / PR Newswire, October 2026] [Citybiz, October 2026].
The substitute markets are also important because they define both upside and boundary conditions. Mach9 does not appear to be selling general-purpose CAD in the Autodesk sense, nor pure GIS, nor raw sensor hardware alone. The product is described instead as a bridge from reality capture into engineering-grade mapping outputs, which places it adjacent to desktop GIS, point-cloud processing, survey drafting services, digital twin workflows, and manual CAD production teams [Forbes, January 2026] [The Geoholics podcast, September 2025] [Informed Infrastructure, March 2026]. If the software works as described, the company can pull spend from outsourced drafting and incumbent geospatial toolchains; if it does not integrate cleanly into those existing workflows, those same adjacent markets become strong substitutes.
Regulatory and macro forces cut in a constructive direction, though the evidence is indirect. Transportation agencies, utilities, and engineering firms operate in environments where documentation accuracy, repeatability, and auditability matter, which helps explain Mach9's reported emphasis on trust, control, and usability at scale rather than autonomous black-box output [Geo Week News]. At the same time, slower public procurement cycles and conservative adoption standards in civil infrastructure can lengthen deployment timelines even when the ROI case is attractive, especially for products that sit close to engineering deliverables and record systems [Morningstar / PR Newswire, October 2026] [TechCrunch, October 2026].
| Cited market signal | Value | What it suggests |
|---|---|---|
| Private beta customers | Nearly 100 | Early demand across a meaningful user base before broad general availability [TechCrunch, October 2026] |
| Projects completed with software | 2,500+ | Usage appears tied to real production work rather than isolated pilots [TechCrunch, October 2026] |
| Surveyed miles by customers | 100,000+ | Indicates infrastructure-scale data volumes, though this is company-reported [Mach9] |
| Total funding after Series A | $49M | Investors see a category large enough to support venture-scale expansion [Morningstar / PR Newswire, October 2026] |
Taken together, the table does not size the category, but it does show where market evidence is strongest: workflow intensity, infrastructure-scale data volumes, and buyer willingness to test automation in production settings. The missing piece is a clean third-party category map for AI-native civil engineering production software, which keeps market precision below what institutional investors would usually prefer.
Partially corroborated -- This section relies on independent reporting from TechCrunch, Forbes, Morningstar / PR Newswire, Citybiz, and Informed Infrastructure, but it does not have a named third-party TAM study for the specific category, and some operating-scale signals are company-reported.
Competition and Substitutes
Positioning
MIXED Mach9 is positioning itself between legacy geospatial and CAD workflows on one side and newer AI-assisted production tools on the other, with the pitch that raw survey and reality-capture data can be turned into engineering-ready outputs inside a more automated mapping environment [TechCrunch, October 2026] [Informed Infrastructure, March 2026] [Mach9].
| Company | Positioning | Stage / Funding | Notable Differentiator | Source |
|---|---|---|---|---|
| Mach9 | AI CAD software for civil engineering that converts 3D and geospatial data into structured maps and engineering workflows | Series A, $49M total disclosed after October 2026 round | Focus on automating conversion of raw survey data into CAD and GIS deliverables, with Digital Surveyor 2 presented as a web-based CAD application | [Morningstar / PR Newswire, October 2026] [Informed Infrastructure, March 2026] [Mach9] |
| TopoDOT | Point cloud processing and feature extraction software used in surveying and mapping workflows | Not established in the provided public sources | Established presence in point-cloud-centered production workflows | [Structured Facts] |
| ArcGIS Pro | GIS desktop software for mapping, geospatial analysis, and data management | Esri product, funding not relevant in this context | Broad geospatial system of record and workflow depth across public agencies and engineering users | [Structured Facts] |
The practical map is narrower than the label "AI CAD" suggests.
Mach9's current wedge appears to be speed at the conversion layer, not ownership of the whole system. Public reporting describes nearly 100 private beta customers and use across more than 2,500 projects, which suggests the company has found real demand among firms that feel the cost of manual drafting and map production every day [TechCrunch, October 2026]. That edge is helped by founder-market fit: the founding team is described as having worked on perception and navigation technology at Tesla, SpaceX, Apple, NASA, and Google X, a background that is relevant when the product depends on interpreting complex physical-world data at scale [Morningstar / PR Newswire, October 2026]. The durability of that edge is less certain, because workflow software tends to become sticky only after it proves trust, accuracy, and repeatability inside customer production environments, and several of Mach9's strongest performance claims remain company-sourced [Mach9] [Geo Week News].
The clearest exposure is that Mach9 does not appear to own the surrounding channels that incumbents already control. Mach9 also looks more exposed in categories where the buyer wants a broad geospatial platform or a formal system of record, rather than a production layer that accelerates map creation [Informed Infrastructure, March 2026] [TechCrunch, October 2026].
The most plausible 18-month scenario is a segmented market rather than a single winner. Mach9 is the likely winner if engineering firms continue to buy around a narrow but painful bottleneck, namely turning large volumes of survey data into engineering-grade deliverables faster and with less manual labor [TechCrunch, October 2026] [Informed Infrastructure, March 2026].
Opportunity
Upside case
PUBLIC The prize here is large if Mach9 becomes the system of record that turns raw geospatial capture into engineering-ready maps and CAD outputs for the firms and agencies rebuilding physical infrastructure, a workflow that still appears fragmented, labor-intensive, and expensive to run at scale [TechCrunch, October 2026] [Forbes, January 2026] [Quiet Capital].
The clearest upside is not simply faster drafting. It is the chance to sit at the conversion layer between reality capture and civil engineering deliverables, where survey data becomes something a transportation agency, utility, or design firm can actually use. Public evidence suggests Mach9 has already moved beyond a concept-stage product into live production environments: TechCrunch reported nearly 100 private beta customers and more than 2,500 projects using the software as of October 2026, while the company and financing announcement name customers that include WSP, HNTB, NV5, Caltrans, Minnesota DOT, and the Maryland State Highway Administration [TechCrunch, October 2026] [Morningstar / PR Newswire, October 2026] [Mach9]. That matters because civil engineering software tends to entrench only after it proves accuracy, trust, and workflow fit, and Mach9's public messaging has increasingly centered on those exact constraints rather than on generic automation claims [Geo Week News] [Informed Infrastructure, March 2026].
A second reason the outcome looks reachable rather than purely aspirational is the company's product direction. Informed Infrastructure described Digital Surveyor 2 as a web-based CAD application intended to help surveyors, engineers, and drafters deliver engineering-grade mapping projects in a single environment, which points to a broader platform ambition than a narrow point tool [Informed Infrastructure, March 2026]. If that architecture holds up in practice, Mach9 could own more of the workflow, capture more seat expansion inside engineering firms, and become harder to displace once a team standardizes around it [Informed Infrastructure, March 2026] [TechCrunch, October 2026].
| Scenario | What happens | Catalyst | Why it's plausible |
|---|---|---|---|
| Multi-office engineering standard | Mach9 starts as an automation tool for survey and drafting teams, then becomes the default production environment across large engineering firms and their regional offices | Digital Surveyor 2's move to a web-based CAD environment broadens adoption beyond a single drafting task and supports shared workflows [Informed Infrastructure, March 2026] | Named users already include WSP, HNTB, and NV5, and the product has reportedly been used on more than 2,500 projects, which suggests repeatable utility inside project-based engineering organizations [Morningstar / PR Newswire, October 2026] [TechCrunch, October 2026] |
| Public-sector workflow foothold | Mach9 becomes a repeat vendor across state transportation agencies and adjacent infrastructure owners that need consistent map production from survey data | Reference customers inside agencies, including Caltrans, Minnesota DOT, and Maryland State Highway Administration, create procurement credibility for follow-on public accounts [Morningstar / PR Newswire, October 2026] | The company already cites agency users, and its stated focus on trust and control aligns with the review-heavy requirements of public infrastructure workflows [Morningstar / PR Newswire, October 2026] [Geo Week News] |
| Geospatial production platform | Mach9 expands from map extraction into a broader system for base maps, line models, and asset records, especially in utility and infrastructure workflows | Product scope already extends beyond raw feature extraction, and company materials describe outputs for electric utilities and infrastructure asset records [Mach9] | The firm's history in perception and navigation, combined with a product built around large-scale geospatial data conversion, gives it a credible starting point for adjacent engineering deliverables rather than a one-feature extension [Morningstar / PR Newswire, October 2026] [Mach9] |
What compounding could look like is fairly specific. Each project run through the platform should improve customer trust in automated extraction, widen the set of accepted deliverables, and create internal references that help one office or agency group sell the workflow to the next. That dynamic is visible, at least in early form, in the progression from point solution language around automated mapping to a broader production-platform framing in company and trade coverage, and in the reported count of projects already run through the software [TechCrunch, October 2026] [Informed Infrastructure, March 2026] [Crunchbase].
There is also a plausible data advantage if Mach9 can learn from repeated infrastructure mapping tasks across roads, utilities, and public works assets, though the public record is still thin on how much of that learning is customer-specific versus cross-customer. The company says customers have surveyed more than 100,000 miles, but that figure remains company-sourced and should be treated cautiously [Mach9]. Even so, if a vendor becomes the preferred layer for converting LiDAR and other 3D inputs into engineering-grade outputs, it can compound through workflow lock-in rather than consumer-style network effects: templates, review conventions, QA habits, and downstream dependencies become hard to unwind once a team is shipping real projects through the system [Geo Week News] [Informed Infrastructure, March 2026].
The size of the win is best framed as scenario analysis, not prediction. Public materials do not provide a defensible market size figure, and no direct public comparable is established in the source set, so the disciplined way to underwrite upside is through workflow ownership and revenue depth rather than a generic TAM claim. If the "multi-office engineering standard" scenario plays out, with Mach9 becoming a deeply embedded software layer across major engineering firms and public-agency workflows, the company could plausibly support venture-scale outcomes associated with vertical infrastructure software leaders, particularly because the October 2026 Series A and $49 million total funding suggest investors see room to build a category platform rather than a services business [Morningstar / PR Newswire, October 2026] [TechCrunch, October 2026]. That is still a scenario, not a forecast, and it turns on whether Mach9 can convert early project usage into durable enterprise deployment.
Partially corroborated -- Built primarily from TechCrunch, Morningstar / PR Newswire, Forbes, and trade coverage, with several important product-scope details still reliant on company materials.
Sources
Open sources
[Y Combinator] Mach9: AI CAD Software for Civil Engineering | Y Combinator | https://www.ycombinator.com/companies/mach9
[Morningstar / PR Newswire, October 2026] Mach9 Raises $34M Series A to Bring Civil Engineering into the AI Era | https://www.morningstar.com/news/pr-newswire/20261021sf76105/mach9-raises-34m-series-a-to-bring-civil-engineering-into-the-ai-era
[TechCrunch, October 2026] Mach9 Raises $34M After Pivoting to AI CAD Software | https://techcrunch.com/2026/10/21/mach9-raises-34m-after-pivoting-to-ai-cad-software/
[The Geoholics podcast, September 2025] Episode 260 - Alex Baikovitz & Mach9 | https://podcasts.apple.com/us/podcast/episode-260-alex-baikovitz-mach9/id1475011444?i=1000725783230
[Informed Infrastructure, March 2026] Digital Surveyor 2 is the next generation of its flagship mapping platform | https://www.informedinfrastructure.com/
[Mach9] Mach9 | https://mach9.ai/
[Geo Week News] Mach9 Unveils an AI-Assisted Workflow Focused on Trust and Control at Scale | https://www.geoweeknews.com/articles/sponsored/mach9-unveils-an-ai-assisted-workflow-focused-on-trust-and-control-at-scale/
[LIDAR Magazine Podcast] LIDAR Magazine Podcast | https://podcasts.apple.com/us/podcast/lidar-magazine-podcast/id1728371618
[Crunchbase] Mach9 | https://www.crunchbase.com/organization/mach9
[Highperformr] Alexander Baikovitz | https://www.highperformr.ai/
[Forbes, January 2026] Mach9 | https://www.forbes.com/profile/mach9/
[LinkedIn, March 2023] Mach9 LinkedIn Profile | https://www.linkedin.com/
[Citybiz, October 2026] Mach9 Raises $34 Million Series A to Expand AI Software for Civil Engineering | https://www.citybiz.co/article/922291/mach9-raises-34-million-series-a-to-expand-ai-software-for-civil-engineering/
[Apple Podcasts] LIDAR Magazine Podcast | https://podcasts.apple.com/us/podcast/lidar-magazine-podcast/id1728371618
[Quiet Capital] Why Data Is the Missing Link for Fixing America's Infrastructure Problems | https://quiet.com/essays/why-data-is-the-missing-link/
Articles about Mach9
- A Carnegie Mellon Spinout Is Building the CAD for America's Infrastructure — The Carnegie Mellon spinout, backed by $49 million, is automating the conversion of raw 3D scans into engineering-grade maps for a critical but slow-moving industry.