The most expensive part of automating a factory line is often the part that never moves. For decades, industrial robots have required custom fixtures, jigs, and dedicated end-effectors to reliably handle parts. This makes sense for stamping out a million identical widgets, but it breaks down completely when the next part on the line is a different size, shape, or material. The promise of flexible automation has been a mirage, chased by expensive, bespoke engineering projects. Prehensio, a 2026 spinout from Germany's Fraunhofer IPA, is betting its entire company on a software layer designed to make that mirage real.
Its core product, HandOS, is pitched as a foundational model for robotic grasping. It generates grasp configurations for industrial robots equipped with dexterous, humanoid-style five-finger hands, allowing them to pick and place a wide variety of parts without manual reprogramming or custom tooling [Prehensio]. The company's entry point is deliberately narrow: solving the grasp-and-manipulate problem for high-mix production environments. The customer value proposition is a reduction in fixtures, programming effort, and deployment friction, not the sale of a general-purpose humanoid robot [WHU].
A software wedge into hardware-dominant automation
Prehensio's strategic wedge is its software-first approach. Instead of selling a complete robot, the company introduces a software layer that operates above existing hardware, continuously adapting execution based on real-time perception, force, and motion feedback [Prehensio]. This allows the system to handle part variability without modifying the underlying robot platform. For a potential buyer, this means the capital expenditure discussion shifts from a massive robotics overhaul to a software integration and the addition of a dexterous hand. It's a pragmatic path to market that avoids the immense cost and complexity of developing a full humanoid system from scratch. The technical origin in Fraunhofer IPA research on grasping variable parts provides a credible foundation for this bet [WHU].
The team and its institutional backing
The founding team reflects the company's hybrid deep-tech and commercial ambitions. The technical core, including CEO Ramez Awad and co-founders Dr. Tobias Schäfle and Nicolai Kilian, emerged from the Fraunhofer IPA research team that worked on the underlying robotic manipulation problems [Prehensio, WHU]. Franz Dornbach, a WHU alumnus, joined as a co-founder responsible for financing and sales, rounding out the early leadership [WHU]. This academic-commercial blend is mirrored in the company's funding. Prehensio secured pre-seed financing from L-Bank, Campus Founders Ventures, and the Fraunhofer Society, in addition to earlier support from Germany's EXIST Research Transfer program [WHU]. While the pre-seed amount is undisclosed, the involvement of these institutional partners, particularly the Fraunhofer Society, signals early validation of the deep-tech approach.
The company's leadership and technical origins are summarized below.
| Role | Name | Key Background |
|---|---|---|
| Co-Founder & CEO | Ramez Awad | Robotic manipulation research at Fraunhofer IPA [Prehensio] |
| Co-Founder & Technology | Dr. Tobias Schäfle | Fraunhofer IPA research team [WHU] |
| Co-Founder & Technology | Nicolai Kilian | Fraunhofer IPA research team [WHU] |
| Co-Founder, Commercial & Finance | Franz Dornbach | WHU alumnus; financing and sales [WHU] |
Where the commercial wheels must meet the road
The transition from impressive research to paid production contracts is the critical, unproven phase. The public record shows technology demonstrations and engagement with potential customers like Müller - Die lila Logistik SE, but no evidence yet of recurring revenue from live production lines [LinkedIn, May 2026]. For a hardware-enabled software company, the risks are multifaceted and must be managed in sequence.
- Proof-of-concept to production. Demonstrating a grasp in a lab is fundamentally different from achieving the required cycle time, reliability, and mean time between failures (MTBF) on a factory floor operating 24/7. The leap involves hardening software for industrial environments.
- The total cost of automation. While Prehensio aims to reduce fixture and programming costs, the dexterous hands themselves are a significant capital outlay. The company must prove its software creates enough operational savings and flexibility to justify the combined hardware and software investment.
- The services burden. Early deployments in complex industrial settings often require heavy professional services. Prehensio must navigate this without becoming a consultancy, building a repeatable, scalable product motion.
The company's most plausible answer to these challenges lies in its focused wedge. By not attempting to solve full humanoid mobility or task planning, it confines the problem space to manipulation, which is still vast but more tractable. Success will be measured by landing a handful of reference customers who can publicly attest to a reduced time-to-automation and a positive return on investment.
The ideal customer and the competitive landscape
Prehensio's ideal customer profile is a manufacturing or logistics operation with high part variability and medium-to-high production volumes, where the cost of custom tooling and constant reprogramming is a known pain point. Think of a automotive supplier handling multiple sub-assemblies, a electronics manufacturer dealing with different product generations, or a third-party logistics warehouse managing diverse SKUs. For these buyers, the benchmark isn't a sci-fi humanoid; it's the total cost and inflexibility of their current automation or manual processes.
The realistic competitive set is fragmented. On one side are the incumbent robotics integrators and automation specialists who build custom solutions, often at high cost and long timelines. On the other are a growing number of AI-for-robotics software startups focusing on specific niches like bin picking or assembly. Prehensio's differentiation rests on combining dexterous hardware with a proprietary software model trained for generalization, a pairing that remains rare outside of research labs. The next twelve months will be about moving from validated research to validated commercial pilots, proving that HandOS can deliver on its promise to make flexible grasping a standard feature, not a custom project.
Sources
- [Prehensio] About the Adaptive Automation Team | https://www.prehensio.ai/about-us
- [Prehensio] Software Tech of Adaptive Automation | https://www.prehensio.ai/tech
- [WHU, February 2026] From Research to Startup: Prehensio’s Robotics Journey | https://www.whu.edu/de/news-insights/whu-magazin/artikel/fraunhofer-ipa-prehensio-robotic-manipulation-startup/
- [LinkedIn, May 2026] Ramez Awad LinkedIn Post | https://www.linkedin.com/posts/ramez-awad-9309863a_robotics-automation-robothand-activity-7449758397069004800-39f1