TABLE OF CONTENTS
Key takeaways
- Legacy PLM systems weren’t built for today’s connected, real-time workflows
- Duro was rebuilt from the ground up with an API-first architecture
- The API is fully open and powers the product itself; not a limited add-on
- A modern GraphQL foundation enables deeper integrations, automation, and AI workflows
- API-first PLM is critical for speed, flexibility, and long-term scalability
- API-first platforms more reliably integrate with AI models to build strong digital platforms
In 2025, we didn’t just refresh our product. Rather, we fully rebuilt Duro from the ground up. I’d like to take a closer look at one of the major focus-points of the rebuild, our API-first architecture, and why it specifically addresses the PLM needs of a modern hardware company.
Modern hardware teams don’t operate in isolation anymore. Engineering, operations, procurement, manufacturing, and increasingly AI systems, all rely on shared product data to move faster and make better decisions. Yet many teams are still constrained by legacy PLM software that act as closed systems, making integrations difficult and change slow.
Traditional PLM solutions were designed for a different era. Integrations were an afterthought, often bolted on years later through limited or proprietary interfaces and heavily dependent on 3rd party consultants. The result is brittle point-to-point connections, manual exports, disconnected and unsupported systems, and teams spending more time managing tools than building products.
At Duro, we chose a different path. When we rebuilt our PLM, we made a deliberate, foundational decision: the API would come first. Not as an extension of the product, but as the core of it. Fully open, fully accessible, and designed to support the way modern hardware teams actually work.
An Open, API-First Foundation
Every core object and workflow in Duro is exposed through the same API that powers the application itself. There’s no separate “internal” API and no artificial limitations placed on our customers. If you can do something in the UI, you can programmatically do it through the API as well.
This is fundamentally different from legacy PLM, where APIs are often read-only, heavily restricted, or locked behind expensive enterprise tiers. In those systems, integrations tend to be fragile and one-off, making them harder to maintain as workflows evolve.
By contrast, Duro’s API is designed to be extended. Customers aren’t forced to integrate around Duro, they build directly on top of it. That architectural choice creates flexibility not just today, but as teams grow and adopt new tools.
A GraphQL API Built for Modern Teams
At the heart of Duro’s platform is a full GraphQL API. GraphQL enables teams to work with product data in a way that matches how hardware is actually structured. Rather than stitching together multiple API calls, teams can traverse complex relationships (like BOMs, revisions, and change orders) in a single query.
This makes it far easier to build internal tools, automate workflows, and connect Duro to other systems without over-fetching data or managing brittle endpoints. Developers can request exactly what they need and nothing more.
To support this, Duro includes an interactive Apollo Explorer that lets developers explore the schema, test queries, and prototype integrations in real time. The experience is intentionally modern and accessible, which is music to the engineer who may have found older PLM systems too inflexible to be of any use.
Supporting Model Context Protocol (MCP) for AI and Automation
AI-assisted engineering isn’t just a flash-in-the-pan at this point. That’s why Duro supports MCP (Model Context Protocol), a standardized way for AI agents to interact directly with systems like PLM. Through MCP, AI models can understand the relationships between parts, assemblies, revisions, and changes; and then take action within workflows.
Instead of building one-off integrations for every new AI tool, MCP provides a consistent interface that allows AI systems to work with Duro as a source of truth. This makes it possible to deploy internal copilots, automate repetitive tasks, and accelerate decision-making without compromising data integrity.
For example, Duro’s MCP can modify components, create a change order, select users to notify, and then connect to Google Workspace’s MCP to create an event on their calendars. Instead of using an API to accomplish this string of tasks, MCPs make it more user friendly for these types of common workflows.
It’s the same mindset behind features like natural language search. Convenience and accessibility in PLM leads to better productivity and lets hardware teams do more of what they want to do, and less of the organizational tasks that we’re taking off their shoulders.
Connecting PLM Across the Entire Stack
PLM software shouldn’t be a data silo. Rather, it should act as the digital thread that connects engineering to the rest of the business.
Because Duro is API-first by design, customers can reliably integrate PLM data with ERP systems, MES platforms, and procurement tools without relying on brittle custom connectors. These integrations are built on a consistent, well-documented API that scales as workflows become more complex.
This enables workflows like automatically syncing approved BOMs into ERP systems, triggering manufacturing processes when engineering changes are released, and keeping supplier data aligned with the latest revisions. Instead of manual handoffs, teams get a single, shared source of truth powering real operations.
Why API-First Actually Matters
Choosing an API-first architecture was a strategic one, both for the Duro team and for our customers. It determines how quickly teams can integrate new tools, how easily processes can evolve, and how well organizations can adopt emerging technologies like AI. An open, API-first foundation enables faster implementation, greater flexibility, and a much more future-proof approach to scaling hardware operations.
Legacy PLM platforms are proving to be far too clunky and inflexible to be as agile as today’s engineering teams. But Duro was rebuilt specifically for the modern scenario, to be an open, API-first PLM designed to support real-world teams in how they work today, and the increasingly automated, AI-driven workflows of tomorrow.
