About yellow3 lab

What happens when people stop operating software and start directing systems?

yellow3 lab is an independent research and venture lab in Copenhagen, Denmark - studying how humans and increasingly autonomous systems work together.

The question

For most of the history of computing, software has been something people actively use. We click buttons, navigate interfaces, fill in forms, move data between systems, and manually execute workflows. Software has largely been a tool that requires continuous human operation.

We believe that is beginning to change.

Advances in artificial intelligence, automation, and digital infrastructure are making it possible for software to take on a growing share of execution. Increasingly, humans define goals, constraints, priorities, and judgment, while systems handle the work required to achieve an outcome.

This shift raises important questions.

These questions sit at the center of yellow3 lab's work.

What we do

yellow3 lab is an independent research and venture lab based in Copenhagen, Denmark. We research how humans and increasingly autonomous systems work together, and how that relationship is reshaping software, organizations, markets, and society.

Our work focuses on the emerging discipline we call Outcome Architecture.

Outcome Architecture is the design of systems where humans define objectives, constraints, and judgment, while software increasingly executes the work required to achieve an outcome.

We believe this represents a fundamental shift in how digital systems are designed and how businesses operate. Just as the internet changed how information moved and cloud computing changed how software was delivered, autonomous systems are changing how work itself is performed.

Our research currently spans areas including:

01 Outcome Architecture
02 Human-Direction AI
03 Autonomous Software
04 Digital Infrastructure
05 AI-Native Markets
06 Compute Sovereignty
07 Organizational Futures
08 Systems & Society

Through essays, frameworks, field reports, and ongoing research, we seek to better understand the forces shaping this transition.

Research and practice

Research is only useful if it survives contact with reality.

Alongside our research activities, yellow3 lab builds and supports companies that put these ideas into practice. These ventures serve as real-world laboratories where concepts are tested against users, markets, regulation, operational constraints, and the complexities of everyday business.

This relationship matters. Research informs practice. Practice informs research. Ideas become stronger when they are challenged by reality, and products become better when they are grounded in clear thinking.

Rather than treating research and execution as separate disciplines, we see them as part of the same process.

Why we exist

Many conversations about artificial intelligence focus on models, tools, and technology. We are more interested in the systems that emerge around them.

Technology matters, but technology alone rarely changes the world. What matters is how people use it, how organizations adapt to it, how institutions respond to it, and how new forms of work become possible because of it.

Our goal is not simply to follow technological change. Our goal is to understand how that change reshapes the structures around us.

We believe the next generation of organizations will look fundamentally different from those that came before. We believe software will increasingly move from being something people operate to something people direct. And we believe understanding that transition will be one of the defining challenges of the coming decades.

We are not trying to predict the future. We are trying to understand how it is being built.
Thomas Christian Melskens
Thomas Chr. Melskens
Founder, yellow3 lab
Copenhagen, Denmark
CIRPASS-2 Consortium Member
European AI Policy Contributor
The founder

Thomas Chr. Melskens is a founder, operator, and researcher focused on how artificial intelligence is changing the relationship between humans, software, and organizations.

Over the past three decades, he has built companies, advised leadership teams, and worked across multiple industries. That experience shaped a simple observation: most business problems are not technology problems. They are decision-making problems, coordination problems, and execution problems.

Today his work focuses on understanding how increasingly autonomous systems change the way those problems are solved.

As founder of yellow3 lab, Thomas researches what happens when people move from operating software to directing systems. His work explores areas including Outcome Architecture, Human-Direction AI, Autonomous Software, Digital Infrastructure, and the future of organizations.

Alongside the lab's research, he actively builds and operates AI-native products. This provides a practical perspective that complements the research. Ideas are tested against real users, real markets, and real operational constraints rather than remaining theoretical exercises.

His interest is not in artificial intelligence as a technology category. It is in the broader question of how people, institutions, and businesses adapt as software becomes increasingly capable of execution.

The work of yellow3 lab is an extension of that question.

Not how to build better tools. But how to build better systems.

Founded
2024
Based in
Copenhagen, Denmark
Entity
yellow3 ApS
Discipline
Outcome Architecture
Work with yellow3

Want to build, understand where AI really stands, or put a hard question to someone who ships?

Let's build what's next - together.

Start a conversation →