Antares Labs Raises $7.25M Seed for Real Estate AI

Antares Labs publicly launched on July 28, 2026, with a $7.25M Seed round led by Fifth Wall. Base10 Partners, Bloomberg Beta, and Sandwith Ventures also participated in the financing.

The Chicago-based company builds custom AI systems inside real estate operators' own environments. Its pitch is less about adding another software interface and more about turning proprietary data, workflows, relationships, and institutional knowledge into systems that can support operating and investment decisions.

The round matters because enterprise AI is moving from broad experimentation toward specific workflows where the buyer controls the data and can measure the result. Real estate is a difficult proving ground for that thesis because its information is fragmented across properties, teams, systems, and years of human judgment.

What Happened

The $7.25M Seed announcement accompanied Antares Labs' public launch. Fifth Wall led the round, with Base10 Partners, Bloomberg Beta, and Sandwith Ventures also participating.

Antares Labs was incubated inside Fifth Wall, an investment firm focused on technology for the built environment. Fifth Wall's sector specialization gives the company proximity to the owner-operators Antares Labs wants to serve, while Base10 Partners and Bloomberg Beta bring stated investment theses around real-economy automation and software that improves how businesses operate.

Noaman Ahmad is the co-founder and CEO of Antares Labs. Before this launch, he served as CFO of Doma and later as CFO of Keystone.

A Different Deployment Model for Real Estate AI

Antares Labs says it begins with the operator's business problem rather than a prepackaged platform. Its forward-deployed teams embed with customers, learn how the business makes decisions, and build company-specific systems around the client's existing data, policies, permissions, and workflows.

That deployment model addresses a recurring enterprise problem: buying AI is easier than integrating it into the decisions that produce revenue, cost savings, or risk reduction. A generic assistant may summarize information, but an operating system must understand which data matters, who can act on it, what controls apply, and how a recommendation connects to a business objective.

Antares Labs says its platform can integrate with AWS Bedrock, OpenAI, Gemini, and Anthropic, then route tasks according to cost, performance, and quality. The company also says customer data and logic remain inside the customer's environment, recommendations are traceable from source data to business strategy, and deployment does not require a new data lake or a wholesale system replacement. These are company-reported capabilities, not independently audited technical findings.

Quarterra Shows the Initial Use Case

Antares Labs names Quarterra as a current customer. Quarterra describes itself as a multifamily development and investment platform and is using Antares Labs to identify and target new submarkets and land parcels.

That example makes the product thesis concrete. Instead of asking a model a general market question, an operator can combine proprietary deal history, internal assumptions, market signals, and institutional judgment to narrow where a team should investigate next. The valuable output is not a clever answer. It is a better decision that fits the operator's actual process.

The company describes additional use cases across lease-up, hospitality operations, industrial inventory, and tenant-churn prediction. Antares Labs also claims customers can see ROI in 30 days rather than 30 months, but no independently audited performance data has been disclosed. That claim should be treated as an execution target until customers publish measurable results.

Why the Investor Mix Matters

The investor group reflects the category Antares Labs is trying to build. Fifth Wall brings built-environment reach, Base10 focuses on automation for the real economy, and Bloomberg Beta invests at early stages in companies that make businesses work better. Sandwith Ventures also participated, although the company did not disclose individual check sizes or board roles.

This is not proof that the company will win, but it does show alignment between the product thesis and the investors' stated areas of focus. Antares Labs is positioning itself at the intersection of PropTech, enterprise AI, and operating transformation, where sales cycles can be slow but successful deployments can become deeply embedded.

The risk sits in the same place as the opportunity. Real estate organizations often have uneven data quality, complex permissions, different systems across assets, and teams whose most valuable knowledge was never written down. Antares Labs must turn that complexity into traceable decisions without becoming a permanent consulting project disguised as software.

What the $7.25M Seed Needs to Prove

The next milestone is not another model integration. Antares Labs needs to show that its forward-deployed approach can produce repeatable outcomes across customers while preserving the company-specific context that makes each deployment valuable.

The most important evidence will be operational: time saved, decisions accelerated, lease-up improved, opportunities identified, or avoidable vacancies reduced. Customer-reported results will also need enough context to separate the value of Antares Labs from the quality of the customer's data and the work of the embedded team.

Antares Labs has a credible opening position: an industry-specific problem, a lead investor with built-world access, a finance-oriented co-founder who understands enterprise decisions, and a named customer using the system for a concrete workflow. The $7.25M Seed gives the company room to test whether AI built inside the business can become more durable than AI layered on top of it.