Startups & Funding

Modern Industrials raises $6M for industrial AI agents

Modern Industrials raised $6 million on November 19, 2026, led by Glasswing Ventures, to scale permission-based AI agents across industrial sales, procurement, inventory and ERP workflows.

Modern Industrials Raises $6M Seed to Expand Custom AI Agents
Modern Industrials Raises $6M Seed to Expand Custom AI Agentsmrpstips / pixabay
By Marcus Bennett5 min read

Why it matters

  • Modern Industrials raised $6 million in seed funding on November 19, 2026.
  • Glasswing Ventures led the round; TMV co-led it; a16z speedrun and other investors participated.
  • Agents remain read-only until customers grant write access scoped to approved actions.
  • Modern Industrials lists at least seven named ERP integrations, including Oracle NetSuite and Microsoft Dynamics 365.
  • Steve Rask forecast savings of hundreds of hours and millions of dollars.

By Marcus Bennett

Modern Industrials raised $6 million in seed funding on November 19, 2026, to expand custom AI agents (software that performs defined tasks inside existing business systems) for industrial manufacturers and distributors. Glasswing Ventures led the round, TMV co-led it, and a16z speedrun and other investors participated.

Chief executive Austin Mao said the capital would accelerate development, expand the team and bring the company’s software approach to more manufacturers and distributors. Modern Industrials builds each deployment around a customer’s operations, with the software working inside systems the customer already uses.

The company targets the manual work connecting emails, spreadsheets and business systems in industrial operations. The focus carries commercial stakes because Modern Industrials says customers can add agent-driven workflows while retaining the enterprise resource planning (ERP) software already used to run their businesses. Customer-specific integration and minimal disruption are central to its scaling plan, according to the company.

Steve Rask, CEO of private equity group AC Summit and MJB Wood Group, forecast that the technology “will save hundreds of hours and millions of dollars.” MJB is a sales, sourcing and manufacturing partner in the wood-products supply chain, and the announcement says it is expanding Modern Industrials to other portfolio companies.

John Chisholm, president of Atlantic Forest Products, said Modern Industrials had found a way to solve each new challenge his team presented. The announcement presents these customer accounts as early examples of the tailored deployments that the seed round is intended to expand.

What workflows are the agents designed to handle?

Modern Industrials designs for industrial and building-materials manufacturers and distributors across five operating areas: sales, procurement, inventory management, operations and customer engagement. According to the company, its named applications and agent functions cover the following work:

  • Quoting and order entry for sales workflows.
  • Demand planning for inventory management.
  • Vendor management for procurement.
  • Customer support for customer engagement.
  • Answering questions, generating insights, making decisions and executing work across sales, procurement, finance and operations.

The company says the objective is to automate complex processes, give teams better information for decisions and minimize disruption to existing operations. Customer-specific integration matters because Modern Industrials designs agents around each customer’s operations instead of treating one workflow as universal.

Modern Industrials starts with the manual work linking emails, spreadsheets and business systems. The company then builds each deployment around those gaps, making integration and approval controls part of the product rather than a later addition.

How does permission-based automation work?

Modern Industrials’ product documentation describes a staged path to automation. Systems remain read-only until a customer grants write access, and that access is scoped to specific approved actions.

Customers decide how much work runs autonomously and can revise that choice later. Most customers begin with people reviewing everything, then allow more automation after the system proves accurate on real order flow, according to the documentation.

Modern Industrials’ documentation separates an agent’s ability to retrieve information, summarize it and make decisions from its authority to execute an approved task. Read-only mode restricts the system to non-writing work. Scoped write access defines which approved actions it can take. The graduated model permits customers to expand autonomy after the system proves accurate on real order flow.

Which enterprise systems can the agents connect to?

Modern Industrials lists integrations with Oracle NetSuite, Microsoft Dynamics 365, JD Edwards, Acumatica, Epicor Eclipse, Epicor Prophet 21 and Infor, among others. According to the company, those connections let customers add agent-driven workflows while retaining the software already used to run operations.

Modern Industrials presents ERP connectivity as part of a gradual deployment model rather than a requirement to replace core systems. Quoting, planning, procurement and support must continue to work while new agent workflows are introduced, according to the company’s stated goal of minimizing disruption.

Modern Industrials says its permission controls let customers govern the pace of autonomy while retaining their operating software. The company’s model therefore links the choice of software with the choice of how much work agents may perform without human review.

Who is behind Modern Industrials?

The three co-founders combine AI engineering and consulting backgrounds, according to the funding announcement. Their experience covers AI strategy, software engineering and machine-learning implementation.

  • Austin Mao, chief executive: Mao worked on Fortune 100 AI strategy and implementation projects at McKinsey.
  • Ankit Bhargava, chief technology officer: Bhargava served on xAI’s technical staff and previously worked as an Uber engineer.
  • Vatsal Bhargava, chief product officer: Bhargava worked as a machine-learning engineer at Google.

Other founding-team members bring experience from Palantir, Meta and J.P. Morgan, according to the announcement. That mix pairs business implementation with AI and machine-learning engineering, while the product remains tailored to each customer’s operations.

What is the investors’ thesis?

Rudina Seseri, Glasswing founder and managing partner, said advances in reasoning and agentic AI (AI systems that can take actions toward a goal) could unlock efficiencies that traditional software could not address. She said the technology had the potential to improve revenue and margins for industrial customers.

TMV co-founder and managing partner Soraya Darabi said the founders’ practice of understanding each customer’s business and building around it helped earn customer trust quickly. Seseri and Darabi’s statements frame the commercial bet around two assets: advances in agentic AI and software tailored to each industrial customer.

The announcement frames the return case in projections. Rask forecast hundreds of hours and millions of dollars in savings, while Seseri identified potential gains in revenue and margins. Modern Industrials still has to convert that thesis into repeatable deployments across its target market.

What must the seed round prove?

The $6 million gives Modern Industrials resources to broaden its model beyond early deployments. Mao tied the proceeds to three priorities: accelerating development, expanding the team and reaching more manufacturers and distributors.

Modern Industrials’ scaling test is whether it can add people and customers while preserving the tailored integrations and graduated, permission-based automation that distinguish its approach. The seed must turn customer-specific deployments into a repeatable commercial model without making permissions broader by default.

Modern Industrials’ next test is whether the $6 million can expand headcount and customer count without losing the bespoke integrations and scoped write permissions underpinning its early deployments. The outcome will show whether agent software can scale across industrial ERP environments without replacing the systems customers already use.

Source: unite.ai

Share this article:

More from Marcus Bennett

Marcus Bennett

Show full bio

Senior reporter covering companies, deals and strategy in artificial intelligence and machine learning at AI In Context.

791 articles

Related articles

  1. IBM and Microsoft Launch Imagine Enterprise for Industry AI Agents
  2. micro1 Pays Upwards of $2 Million for Operational AI Data
  3. Model ML raises $250M Series B at $1.5B valuation for finance AI agents
  4. Islo Raises $15 Million for Controlled AI Software Factories
  5. Riste Gramatikov Maps Palantir’s Forward-Deployed AI Model

« Previous articleNext article »