Across the United States, the push to bring more AI capacity online has created a construction boom unlike anything in recent memory. Data centers are being planned in rural counties, clean-energy facilities are moving through permitting, and manufacturers are expanding existing plants. Each of those projects brings its own engineering challenges, but there is another challenge that often gets less attention: compliance. The rules governing these projects can be layered, contradictory, and expensive to ignore. Dili, a startup that emerged from Y Combinator, is aiming to solve that problem with an AI platform designed specifically for construction compliance.

On Thursday, Dili announced a $15 million Series A round, bringing the company’s total capital raised to $21.7 million. The round follows a $6.7 million seed round and positions Dili to expand its platform beyond its current base of customers. The company’s software is already being used on about 700 projects, according to co-founder and CEO Anand Chaturvedi, and the new funding will help the company scale its operations and refine its compliance engine.

An unusual bet on infrastructure compliance

The Series A was led by Khosla Ventures, a firm known for backing ambitious artificial-intelligence companies. Participating investors included Allianz, Rebel Fund, Brick and Mortar Ventures’ Darren Bechtel, and Y Combinator’s Garry Tan. Dili was part of Y Combinator’s Summer 2023 batch, giving the startup access to a broad network of founders and operators in the construction and energy sectors.

“AI for compliance” has become a common pitch among startups, but Dili focuses on a narrow and particularly messy slice of the market: the rules that govern construction projects, especially those receiving some form of federal funding. That focus is deliberate. The infrastructure boom tied to AI data centers, the Inflation Reduction Act, and reshoring of manufacturing has created an urgent need for tools that can handle wage rules, safety regulations, and environmental requirements at scale.

Why compliance is so difficult

To understand the problem, consider the rules that apply to a single federally funded project. Under the Davis-Bacon Act, the Department of Labor sets prevailing wages for workers on certain public works jobs. A separate set of prevailing wage and apprenticeship rules, often called PWA rules, applies to clean-energy projects funded under the Inflation Reduction Act. Additional OSHA workplace-safety requirements and EPA environmental rules may apply depending on the nature of the work. Each of these regimes has its own definitions, reporting schedules, and penalties.

• Davis-Bacon requires contractors to pay wages set by the Department of Labor for specific trades and regions. The requirement extends to subcontractors and often covers projects that receive federal funding.
• PWA rules add apprenticeship quotas and certification standards, typically requiring a certain percentage of labor hours to be performed by registered apprentices.
• OSHA and EPA rules vary by project type and can impose separate record-keeping and inspection obligations.

The overlapping requirements make manual compliance difficult. A compliance team might normally sample a subset of payroll records, invoices, and apprenticeship documents, hoping to catch problems before they become costly. Dili, by contrast, checks all the information as it comes in. “Non-compliance can result in millions of dollars of fines for those projects,” Chaturvedi explained. “So it’s really powerful to be able to check all the information as it comes in, instead of just sampling data.”

Keeping AI in its lane

Given those high stakes, reliability is a central concern. Chaturvedi believes Dili’s architecture prevents any LLM-based fuzziness from sneaking into the final product. Contemporary AI models are only used in the company’s data layer, the engine that takes unstructured documents and translates them into structured data. From there, a deterministic system sorts the data according to the complex-but-static compliance rules. The distinction is important for customers who need an auditable, repeatable answer for regulators.

“Imagine being able to read across the entire context of a company’s internal documents, all of their vendors’ documents, all of their ERP information, all of their payroll systems information, and then draw out the data that you need specifically for reporting or compliance,” Chaturvedi said. The AI layer is designed to make that possible without forcing every subcontractor onto a single software platform. If a vendor sends a PDF, a spreadsheet, or even a scan of a physical document, the data layer can extract the relevant fields.

The deterministic rule engine then applies the legal requirements. For example, if a worker’s classification is pulled from a payroll record, the engine can check it against the prevailing wage schedule for that trade and location. If the wage is too low, the system flags the discrepancy immediately. If an apprentice certification is missing, the project manager sees the issue before a regulator does. This approach turns a compliance task that might take a full day of manual review into a process that can be completed in minutes.

Infrastructure boom and the compliance bottleneck

The need for such a system has grown alongside the infrastructure boom. AI data centers require massive amounts of power, which has led to a surge in new energy projects, transmission lines, and battery-storage facilities. Many of those projects are built with the help of federal tax incentives, meaning the contractors must meet the labor standards attached to those incentives. At the same time, reshoring and supply-chain diversification have pushed manufacturers to break ground on new plants, many of which also rely on federal support.

Construction has historically been slow to adopt digital tools. Payroll records arrive in different formats from different subcontractors, and a general contractor often has to reconcile paper documents with ERP systems and union reports. Dili’s technology is built for that patchwork. By making the data layer AI-driven, the company can handle variation in document types and formats without forcing every subcontractor onto a single compliance suite.

Deployment and business model

Dili is already making the system work in practice. Chaturvedi said the software is being used at about 700 projects, ranging from manufacturing facilities to data centers. Notably, roughly half of those projects use Dili as an in-house software tool, while the other half outsource the entire compliance process to the company on a contractor model. Dili is able to handle both types of contract, although Chaturvedi anticipates the industry will shift more toward the software model in the years to come.

“Software and AI are going to start eating a lot of those professional services workflows, so I think more and more people will start to bring those in-house,” Chaturvedi said. “The interesting thing will be how the market itself evolves and where the customer needs go as AI develops.”

The company’s growth suggests that the market is ready for that shift. Between the seed round and the new Series A, Dili has raised $21.7 million, with strong support from investors who have deep ties to both AI and industrial construction. The company’s Y Combinator pedigree and its focus on a concrete, high-value problem give it a clear path as federal infrastructure spending continues to expand.

For project owners, contractors, and subcontractors, the takeaway is that compliance is becoming an automated process. The tools are still early, but the trajectory is clear: AI will be used to extract facts from messy documents, and deterministic software will apply the rules. That combination is already reducing the time and cost of compliance on hundreds of projects, and it is likely to become the standard for the next generation of infrastructure development.

Source:TechCrunch News