How Ursa Major Combines EOS Open APIs and AI to Redefine Manufacturing
16 October, 2026 | Reading time: 5 min
Manufacturing speed has always been a competitive advantage, but in the defense and aerospace sectors, it’s often a matter of national security. As the U.S. defense industrial base faces mounting pressure to replenish depleted munitions and scale hypersonic capabilities, traditional manufacturing pipelines are proving to be major bottlenecks.
Ursa Major, an aerospace and defense company based in Berthoud, Colorado, is tackling this challenge head-on. By pairing industrial metal additive manufacturing (AM) with advanced artificial intelligence and open software architecture, Ursa Major is proving that the future of defense production isn't just about faster lasers - it's about smarter, software-driven iteration.
Software-Driven Propulsion: From Hypersonics to Flexible Solids
Ursa Major’s mission spans three critical defense pillars: hypersonics, in-space propulsion, and solid rocket motors. Across all three, AM serves as the primary engine for rapid innovation and hardware execution.
In hypersonics, Ursa Major’s flagship Hadley engine has successfully flown over nine hypersonic missions. It’s an oxygen-rich staged combustion liquid rocket engine that is roughly 80% 3D-printed by mass. Partnering with Shadow Launch, the team has repeatedly recovered, inspected, and re-flown the exact same printed engine more than nine times. Simultaneously, their storable liquid engine, Draper, recently completed successful test flights in a joint program with the Air Force Research Laboratory).
While liquid engines leverage AM for extreme design complexity and thermal management, solid rocket motors present a different challenge: supply chain flexibility and cost. Traditional solid rocket motor production relies on rigid, multi-million-dollar fixed manufacturing lines. If a defense program requires a different motor diameter, building a new line can take years and hundreds of millions of dollars.
Through its proprietary LINX manufacturing process, Ursa Major uses a flexible "unit cell" approach. Utilizing EOS metal 3D printing systems, the team can produce multiple different-sized motor cases and complex internal tooling on the exact same machine. By replacing long-lead forged components, which often carry 12-month lead times for exotic steels, with printed hardware, Ursa Major can design, print, test, and iterate in weeks.
Vibe-Coding and AI Acceleration
Building flexible hardware requires equally flexible software. This is where AI and custom software development are creating a step-change in manufacturing productivity.
Using AI-assisted development tools - often referred to as "vibe-coding" - Ursa Major engineers integrate large language models into their custom slicing repository (Polaris). Rather than spending days manually configuring complex designs of experiments (DOEs) or writing low-level scripts, engineers use natural language prompts to rapidly prototype and test advanced algorithms.
For example, an engineer can feed a newly published academic paper on laser scanning strategies into an AI coding assistant, instruct it to write a new scanning function within the slicer repository, and apply custom parameters to individual parts on a build plate.
In a recent demonstration, Ursa Major generated a part-specific thermal management strategy that sorted hatch tiles from the outside-in based on each component's centroid. The AI tool updated the configuration files, built the required support structures, and prepared the slice file for an EOS M 400 platform. The upshot: days of manual programming became a 10-minute task.
The Power of System Openness
AI algorithms are only as effective as the data streams they can access. To build true end-to-end automation, software tools must seamlessly interface with physical machinery.
This is where EOS sets the industry standard in openness, flexibility, and machine accessibility. Unlike closed ecosystems that restrict user access, EOS provides robust, fully documented application programming interfaces, including the EOS REST API and EOS CONNECT.
Using these open interfaces, Ursa Major built a fully customized, web-based fleet monitoring system in just one week using 100% AI-generated code. The system queries EOS machines rapidly and repeatedly, logs build data into a local SQL database, projects live camera feeds across active printers, and sends automated status reports. If an anomaly occurs - such as a short feed or recoater snag - the system immediately flags the issue and emails engineers a real-time image of the build chamber.
"The real unlock in the future is going to be the companies that have really good documentation for how these tools interface with the real world or with hardware.” He went to cite how he used the EOS CONNECT Core API. “All I did was take the documentation, feed it into CloudCode, and say, 'Hey, here's all the documentation for this. What we wanna do is build a database that can query the machines every minute, every thirty seconds, whatever, and then pull the data off … And it can do all that on its own without any prompting.”
Tom Pomorski, Director of Additive Manufacturing at Ursa Major
By providing transparent system access, EOS enables users to build custom data fusion pipelines, connecting serialized part tracking, sensor logs, and machine diagnostics directly into enterprise manufacturing systems.
Bridging Development to Production
While AI-driven vibe-coding is revolutionary for rapid prototyping and process development, transitioning hardware to flight-critical production still demands rigor and precision.
Neural networks and generative scripts can rapidly explore edge cases, but qualified aerospace and defense parts require deterministic verification, closed-loop process validation, and strict quality control. AI accelerates the discovery phase, allowing software teams to refine requirements and hand off proven frameworks to software engineers who build production-ready, enterprise-grade software.
This reality highlights the dual role EOS plays in the AM ecosystem. EOS serves high-capability companies like Ursa Major who require open APIs and deep parameters to build bespoke digital pipelines. Simultaneously, EOS delivers validated, out-of-the-box software and hardware solutions for industrial customers who need immediate, standardized production.
The Future of Advanced Manufacturing
The convergence of multi-laser hardware, open API architecture, and AI-assisted software is reshaping what’s possible in aerospace and defense manufacturing. By removing software barriers and embracing machine accessibility, modern manufacturers can build agile, scalable production cells that respond dynamically to national security needs.
To explore how open software architecture can elevate your production line, learn more about EOS Scalable 3D Printing Software for Productivity.
