Fil Aronshtein on DRA's maritime deal and selling into defense-adjacent industrial markets

Summary

Fil Aronshtein, founder of DRA, has closed a maritime deal — the company's newest vertical alongside airspace, automotive, agri, and construction — though the formal announcement is still pending.

The pitch DRA is making isn't primarily about AI. It's about a workforce crisis that is largely missing from the policy conversation. The average age of the American manufacturing worker is 55, and as that generation retires, it takes with it decades of undocumented tribal knowledge — process specs, workarounds, and institutional memory that never made it into any system. Aronshtein puts it bluntly: the US has spent billions on environmental conservation and zero on tribal knowledge conservation.

The AI nuance

Aronshtein's view on AI in manufacturing is more skeptical than the conference narrative might suggest. General-purpose LLMs were trained on the open internet — forums, documentation, public text — not on the private, proprietary specs and workflows that define actual shop-floor production. That knowledge was never written down, and even where it was, it lives in private databases that were never crawled. For AI to meaningfully disrupt manufacturing, it has to be embedded directly in the workflow, fed the right context, and accessible to workers who are not technically sophisticated. Off-the-shelf LLMs don't get you there.

On the specific question of why American manufacturers are slow to respond to inbound inquiries — a data point circulating that US shops take a week to reply while Chinese counterparts respond immediately — Aronshtein frames it as a culture problem rather than a technology problem. Companies like Paperless Parts and Xometry have addressed the quoting software problem for years; it's computationally complex geometry, not an LLM gap.

The competitive framing

Aronshtein argues the US manufacturing lag against China isn't primarily a technology gap — it's a hunger gap. Post-Cold War, the US offshored production to China, Japan, and Korea under the assumption that globalism would hold and those relationships were stable. China's current posture, in his framing, is the equivalent of 1776: a generation of underdog manufacturing culture built up over 30 years, now refusing to stay in a subordinate position.

The labor argument

The goal Aronshtein advocates isn't cheap labor — it's smart labor. Technology adoption in manufacturing has historically followed a top-down pattern: Lockheed, Toyota, and Ford can buy the best tools, while the broader industrial base gets left behind. Saving the American industrial base means bringing the bottom of that distribution up. The aspiration is to make manufacturing facilities cool enough that younger workers choose them, the way software engineers are drawn to companies with sophisticated internal tooling. He points to SpaceX's internal manufacturing software — known internally as Warp Speed — as an example of how good tooling creates a talent flywheel, with several former SpaceX engineers having already spun out companies based on that platform.