Antares Labs has launched publicly with $7.25 million in seed funding to build company-specific artificial intelligence systems for large real estate owners, developers, investors and asset managers. The round included participation from Fifth Wall, Base10 Partners, Bloomberg Beta and Sandwith Ventures.
Antares Labs plans to use the financing to expand its platform and forward-deployed teams, which work directly inside customer organizations to understand their operating processes and build AI systems around specific business priorities.
Rather than offering a general-purpose dashboard or software product that customers must configure independently, Antares Labs begins with an operational problem and develops an AI system around the customer’s data, workflows, relationships and institutional knowledge.
The company is targeting a longstanding challenge across the real estate industry. Large owners and operators frequently possess decades of information about acquisitions, leasing, tenant behavior, property operations and market conditions, but much of that knowledge remains distributed across spreadsheets, databases, documents and individual employees.
Antares Labs is designed to convert that fragmented information into AI systems that can support decisions and perform ongoing work inside the organization.
CEO Noaman Ahmad said real estate businesses increasingly want AI that can produce measurable results rather than limited pilots, reports or standalone tools requiring additional management.
The company’s forward-deployed model places technical and industry specialists alongside customer teams. These employees learn how the business operates, identify high-value workflows and create tailored AI capabilities that fit existing systems and decision processes.
This approach is intended to reduce the need for a customer to build a large centralized data lake, replace its current software or complete a multiyear technology transformation before receiving value.
Antares Labs said customers can begin seeing a return on investment within 30 days, although actual results will depend on the use case, available data and implementation scope.
The company’s platform can connect with major foundation model providers, including Amazon Web Services Bedrock, OpenAI, Google’s Gemini and Anthropic.
It dynamically routes requests among models based on factors such as cost, performance and output quality.
This multi-model architecture is intended to prevent customers from becoming dependent on a single AI provider. It also allows Antares Labs to select different models for different tasks as capabilities and pricing change.
One model may perform better when extracting information from lease documents, while another may be more suitable for forecasting, summarizing operational data or generating recommendations.
Because Antares Labs deploys its technology inside the customer’s environment, the organization retains control of its proprietary data, operating logic and AI workflows.
This structure can be important for institutional real estate companies that handle confidential financial information, tenant data, investment strategies and internal underwriting assumptions.
The company said recommendations are fully traceable from the underlying data through the identified signal and resulting business strategy.
That traceability is designed to support review by finance teams, compliance personnel, senior executives and boards of directors.
AI systems used in investment and operating decisions must often explain why a recommendation was produced. A prediction without a clear connection to the underlying information may be difficult for executives to trust or approve.
Antares Labs seeks to address this concern by making its systems auditable rather than operating as a black box.
Its potential applications span real estate investment, development, leasing and property operations.
For a multifamily developer, the platform could analyze market, demographic and property information to identify submarkets that may offer stronger future returns.
This could help an investment team prioritize opportunities before attractive locations become widely recognized or competition increases.
For hospitality real estate investment trusts, Antares Labs can connect real-time market information with operational decisions involving staffing, service levels and spending.
Hotel owners often depend on operating reports that show performance after the relevant period has ended. A more continuous system could help them respond earlier to shifts in demand, pricing, labor needs or customer behavior.
Across different property categories, Antares Labs can also help predict which tenants are more likely to leave.
Tenant turnover can create lost rental income, leasing costs, renovations and extended vacancy periods. Earlier identification of churn risk could allow an owner or asset manager to engage the tenant, adjust terms or prepare a replacement strategy before the lease is lost.
The company’s broader objective is to build what it calls an AI brain for real estate.
This refers to an evolving system that captures organizational knowledge and becomes more useful as it processes additional data, decisions and feedback.
Instead of treating every analysis as an isolated task, the platform can retain information about the company’s objectives, investment criteria, operating practices and prior outcomes.
Antares Labs said that accumulated intelligence can help its systems improve over time while remaining tailored to the individual customer.
The company is already working with major real estate owner-operators, including Quarterra.
Fifth Wall founder, CEO and Chief Investment Officer Brendan Wallace said large real estate companies have accumulated valuable information and relationships that have not been fully activated through technology.
He believes organizations need systems that continuously support their strategies rather than relying only on data stored in applications that employees may not use effectively.
Ahmad said Antares Labs will focus on strategic capabilities that can directly affect investment performance, lease-up and property operations.
The company is positioning its offering as embedded intelligence owned by the customer rather than another outside software vendor that must be managed.
Antares Labs’ launch comes as real estate institutions increase their investment in AI while evaluating how the technology can produce operational and financial results.
The company believes its combination of embedded teams, model flexibility, customer-controlled deployment and industry-specific workflows can help real estate businesses move from experimentation into everyday AI use.
KEY QUOTES:
“Every real estate owner-operator is asking the same question right now: how do we actually put AI to work inside our business? Not a pilot, not a dashboard, something that runs on our data, solves our problems, and compounds over time.”
“We embed directly with our clients to build AI around strategic capabilities that move the needle, accelerating investments, improving lease-up. The output isn’t a report or vendor to manage. It’s intelligence that lives inside their business, owned by them, and gets smarter daily.”
Noaman Ahmad, Chief Executive Officer of Antares Labs
“For over a decade, we’ve heard the same frustration from the largest owner-operators in the world, decades of institutional knowledge, proprietary data, and hard-won relationships that have never been fully activated.”
“Antares Labs was built to be that for real estate operators. The window to move is now, and the cost of waiting is real.”
Brendan Wallace, Founder, CEO and Chief Investment Officer of Fifth Wall
