Aixelo’s Edison-4.0™: Bringing Human-in-the-Loop AI to R&D

At the 5-HT Innovators Club on 10–11 September 2025 at KIT Karlsruhe, Aixelo presented how adaptive, human-guided AI can become part of everyday laboratory practice.

In this 5-HT Startup Story, CEO Dr. Christoph Kreisbeck explains how Edison-4.0™ helps chemical and pharmaceutical R&D teams move beyond trial-and-error, accelerate experimentation, and scale AI across their organizations.

Aixelo CEO Dr. Christoph Kreisbeck presenting at the 5-HT Innovators Club on 10–11 September 2025 at KIT Karlsruhe

How would you explain your solution in three sentences to a professional colleague in a chemical/pharmaceutical company or other potential client?

Aixelo’s Edison-4.0™ is an enterprise-ready R&D management and Materials AI software platform for materials and pharmaceutical research teams. It enables scientists to integrate adaptive, human-in-the-loop optimization directly into their routine lab workflows, replacing trial-and-error with smarter experimentation. Edison-4.0™ has demonstrated 2–10× productivity gains, helping, for example, a partner turning a stalled six-month project into success within six weeks at 75% lower cost.

Aixelo’s Edison-4.0™ software platform

What problem motivated you to start the company?

Our journey began at the intersection of two paths. Manuel, leading a lab at KIT, saw that AI could revolutionize experiment planning for his projects but struggled to find ready-to-use AI tools. Christoph, coming from an AI startup out of Harvard, noticed the same challenge in industry: AI is advancing fast, but deploying it in lab operations lags.

The usual solution: collaborating with AI experts. This is what Manuel did, collaborating with Pascal, Professor in materials AI at KIT. But this approach is not scalable for broader roll-out of AI. Feeling this pain ourselves inspired us to found Aixelo: to bridge the gap, make AI accessible to scientists, and transform R&D across industry.

Aixelo Founding Team Dr. Christoph Kreisbeck, Dr. Manuel Tsotsalas, Dr. Pascal Friederich (from left to right)

How do you convince a chemical/pharmaceutical company or another potential client to set up a pilot project with you?

When chemical or pharmaceutical teams explore AI, the challenge isn’t whether it works, they’ve already seen numerous proof points. The real question is how to move AI from concept to lab operations. Many lab teams are unsure where to start, worried about messy data, automation, or tracking all metadata.

Our first advice is simple: don’t worry about these details. Bring us your most important and challenging project. Do you need to navigate many design parameters, balance multiple objectives, does it take lots of lab resources? These are exactly the projects where anything smarter than trial-and-error delivers real value.

From there, the journey is straightforward. In our onboarding session, teams define their experimental setup, parameters, and objectives, often sparking deeper discussions about the science. Then, Edison-4.0™ proposes experiments, learns from results, and replicates the natural “test-learn” cycle, while uncovering hidden patterns in complex data.

What sets us apart is our holistic, human-in-the-loop approach. We optimize not just single experiments but entire multi-step workflows, from formulation to coating to devices. Unlike data platforms, Edison-4.0™ continuously learns from experiments, guiding scientists through complex optimization while reducing adoption barriers and delivering measurable impact. By the end of the pilot, teams gain not just new results, they acquire a new way of managing R&D projects to scale AI from pilot to full lab operations.

Who are your current customer (groups) and who do you want your potential customers to be?

Our system is general. Edison-4.0™ can address specific lab tasks, such as material synthesis, formulation development, and device optimization, but can also apply to multi-stage R&D processes.

We focus on three customer segments:

• Midsize companies: Often lack in-house AI teams or large budgets; Edison-4.0™ accelerates AI adoption and reduces trial-and-error without heavy investment.

• Large corporations: Need scalable AI solutions and a partner to deploy ready-to-use AI tools across multiple labs.

• Lab equipment providers: Benefit from self-optimizing instruments and integration of Aixelo’s AI experiment planner with automation, enabling off-the-shelf self-driving lab systems.

Where do you see yourself in 3 years and how can 5-HT support you?

In three years, Aixelo will be a reliable AI partner for chemical and materials R&D teams. With Edison-4.0™, scientists gain a reliable companion for AI-driven experiment planning. Our platform streamlines project management from single experiments to cross-lab pipelines, reducing friction and ensuring AI tools are used where they have the most impact. 5-HT can catalyze this journey by connecting us with industry partners, enabling pilots, and providing a network of innovators.

You participated in the 5-HT Innovators Club in September 2025. What is your impression of the event and what have you gained (so far) from participating?

The 5-HT Innovators Club was an excellent opportunity to connect with peers and explore industry trends in lab digitalization, automation, and self-driving labs. It was invaluable for us to hear from leading industry players, not only about the opportunities but also the challenges of deploying AI as a standard in R&D processes.

The day was full of insightful discussions. Of particular value were conversations with in-house data science teams from large corporations. Learning about their tools, workflows, and perspectives on future roadmaps was extremely valuable and has already led to follow-up discussions and potential collaboration opportunities.

Aixelo CEO Dr. Christoph Kreisbeck presenting at the 5-HT Innovators Club on 10–11 September 2025 at KIT Karlsruhe