Synera names Ben Hesseldieck head of engineering to scale its agentic AI in design at Airbus, BMW and NASA

By Zendoric · October 1, 2026.

Synera, a German startup founded in Bremen in 2018, has appointed Ben Hesseldieck as head of engineering, the company announced. Hesseldieck will lead software development and the platform's technical strategy as Synera attempts to scale its AI agents in engineering operations across automotive, aerospace, aviation and defense. He was the fifth employee and one of the founding engineers at n8n, the well-known open-source workflow automation tool, and arrives after a stint at Code & Co as vice president of technology, where he audited the technology of growing software companies.

What exactly does Synera do? According to the company, its platform already runs more than 170,000 automation workflows and has close to 30 agentic engineering projects underway with Tier 1 suppliers, prime contractors and major original equipment manufacturers. Its customers include NASA, BMW and Airbus, as well as six of the world's ten largest carmakers by revenue. The platform connects with more than 90 CAx tools (computer-aided engineering, the software used to design parts and systems), PLM (product lifecycle management, which syncs technical data from design through maintenance) and ERP (enterprise resource planning, the systems that handle procurement and production), without forcing companies to replace that software.

Daniel Siegel, Synera's chief product officer, sums up the bet with a distinction worth keeping: "engineering is splitting into two layers. There's the requirements layer, where engineers decide what to build and why. And there's the execution layer, where data is adjusted by hand between different tools." By the company's own estimates —a self-reported figure with no known external audit— nearly half of an engineering team's capacity now goes to that repetitive execution layer. Hesseldieck frames it in the same terms: "R&D engineering teams still spend too much time on repetitive work that diverts needed capacity and doesn't fully leverage their expertise."

His immediate priority will be Synera Assistant, currently in closed beta and slated for release in early 2027: a tool that will allow users to query product knowledge in natural language, build their own agents, generate 3D designs and map processes. The stated longer-term goal —still aspirational, not a demonstrated capability— is for the system to interpret the engineer's intent, generate precise 3D geometry and operate CAx software directly to produce designs that are increasingly production-ready. Synera, which already has a presence in Boston and San Francisco in addition to its German headquarters, is using this hire to reinforce precisely that transition.

The news matters less for the appointment itself than for what it confirms about where agentic AI is heading: not just writing code or answering emails, but operating the systems that underpin the physical manufacturing of cars, planes and satellites. It is far less flashy terrain than consumer chatbots, but with a wider industrial impact, because it touches supply chains where a design error can't be fixed by reloading a page. Hiring n8n's founding engineer in particular is no accident: his experience scaling an automation software organization is the kind of operational scaffolding that separates a pilot with a one-off customer from a platform that truly works across dozens of manufacturers at once.

This connects with something we have been pointing out sector by sector: the first thing AI displaces is not engineering judgment, but the administrative, repetitive work that surrounds it —adjusting dimensions, syncing versions between CAD and ERP, moving data between systems that don't talk to each other. If Synera itself is right and nearly half of engineering capacity now goes into that execution layer, the short-term promise is freeing up time for design; the risk, just as real, is that draftsmen and engineering data technicians see their roles shrink before demand for judgment-based profiles absorbs those who hold those jobs today. It is the same hard transition we have documented in other sectors, now applied to physical engineering.

In the long run, the same shift points to the underlying thesis we defend at Zendoric: if agentic AI truly manages to absorb the execution layer of physical engineering, what gets freed up is human time for the design, innovation and judgment that determine whether a plane flies safely or an electric car lasts its service life. But it is worth not confusing traction with achievement: 170,000 active automation workflows are a real signal of adoption, while agents capable of generating production geometry without supervision remain, for now, a goal for 2027 and beyond.