How to Become an Art Director as a 3D Artist

A conversation with Dong-Uk Kim

Dong-Uk Kim works at the intersection of 3D, art direction, technology, and increasingly, AI. His move into art direction wasn’t a sudden career switch, but a gradual shift from asking how to make an image to questioning why it should exist in the first place. We spoke with him about developing taste beyond technical skill, staying hands-on as a director, and why the ability to decide what is worth making may become increasingly important as the boundaries between creative tools continue to disappear.

You started as a 3D artist and now work as an art director. How did that transition happen?

I don’t think there was a single moment when I suddenly switched from being a 3D artist to becoming an art director. It felt more like the scope of my work gradually expanded.

At first, my role was mainly about translating an existing idea into strong 3D imagery. But as I worked on more projects, I became increasingly interested in what happens before that stage: why we are making a certain image, what kind of image is most appropriate for the project, and what kind of tone, movement, and atmosphere it should have.

The same thing happened in my personal work. I became less interested in simply making something technically new, and more interested in building a world or a mood and making sure every image within it felt consistent.

Eventually, that naturally led me to take on more responsibility in client projects as well: proposing references, shaping concepts, and defining the overall visual direction. So for me, it was less about changing positions and more about gradually widening the way I look at a project.

Did this “switch of positions” change how you approach creative work? Do you still work as a 3D artist as well?

Yes, it changed my approach quite a lot.

In the past, when I saw a new project, my first thought was usually, “How can I make this?” Now I tend to start with questions like, “Why should we make this?” or “What should people feel when they see it?”

The technical part comes after that.

And I still work directly as a 3D artist. In fact, the process of making things myself is still very important to me. When I build a simulation in Houdini or test a look, unexpected results often appear, and those discoveries can influence the direction of the project again.

So I don’t really separate art direction and production completely. I think being someone who still makes things with my own hands is an important strength in the way I direct work.

What does your work as an art director look like now?

It depends on the project, but I usually start by defining the core impression or message the project should communicate.

From there, I gather references and begin building a visual language around things like tone, color, material, camera, movement, and scale.

Then I keep testing and refining those choices throughout production so that the original direction doesn’t get lost.

I spend a lot of time looking at images in the early stages of a project. Finding good references is important, but I think it is equally important to decide what the project should not be.

For me, art direction is often less about continuously adding new ideas and more about removing possibilities until only the things that are truly necessary remain.

On the Centre Pompidou Hanwha project, you moved from creating images to shaping the overall visual direction. What did that process look like in practice?

On this project, the direction didn’t come from defining one big concept at the very beginning. We found it through a series of visual experiments.

I initially created image and motion tests myself, and through that process we compared different forms, movements, and visual approaches to understand what worked best with the space.

Because the work was designed to be experienced in a physical environment, it was important to think beyond what looked beautiful on a computer screen. We had to consider the scale of the architecture, the viewing distance, and the way people would encounter the imagery in the actual space.

Once a few key visual rules began to emerge, the focus shifted from creating individual images to making sure all of the content felt like it belonged to the same world.

We continued adjusting elements such as color, form, density, and the speed of movement until the overall visual identity became coherent.

In the end, I saw the project less as an exercise in making individual images and more as an attempt to make multiple images and movements feel like one continuous experience.

What do you think 3D artists can do to develop stronger taste and creative direction beyond improving their technical skills?

I think one of the most important things is to look beyond 3D.

As 3D artists, it is natural for us to look at a lot of work made by other 3D artists. But if we stay within that circle for too long, we can easily end up referencing the same aesthetics and the same ideas.

I think it is very important to look at photography, cinema, architecture, fashion, graphic design, contemporary art, and other disciplines.

And rather than simply collecting images you like, it is useful to ask yourself, “Why do I like this?”

Is it the color? The composition? The unusual proportions? The lighting? Or maybe some kind of tension that is harder to explain?

By repeatedly analyzing those reactions, your own taste starts to become more specific.

Technical skills can be learned in relatively clear steps. Taste, on the other hand, is something that develops over a long period of looking, choosing, and rejecting.

Would you please be so kind and share the top 5 of the biggest inspirations for your work?

Rather than choosing five specific artists, I would probably describe five things that repeatedly influence my work.

1. Nature and physical phenomena
   I am very interested in water, smoke, particles, wind, light, and other physical phenomena. I especially like movements that cannot be completely controlled.

2. Architecture
   I am very interested in the relationship between imagery, space, and scale. When I look at architecture, I think not only about its form but also about how people experience the space around it.

3. Photography and cinema
   I am constantly inspired by the way camera position, lenses, and light can transform an ordinary subject into something completely different.

4. Contemporary art and design
   I often look at how a simple idea can be communicated with clarity and strength, rather than focusing only on the final image itself.

5. Technology
   New technology is also a major source of inspiration for me. From tools like Houdini to generative AI, whenever a new tool appears, I start thinking about what kinds of images or experiences might now be possible that were difficult or impossible before.

How are you perceiving the 3D and creative industry right now? Where do you see its biggest limitations, and where do you see the most exciting opportunities?

I think we are in a very confusing but also very exciting period.

Technically strong images are becoming easier to produce. 3D tools are becoming more powerful, and AI is rapidly reducing the time and cost required to generate imagery.

Because of that, I think it will become increasingly difficult to differentiate yourself purely through execution.

At the same time, the ability to decide what should be made, and to connect multiple technologies around a strong idea, will become more valuable.

What excites me most is how quickly the boundaries between 3D, real-time graphics, AI, and interactive technologies are disappearing.

In the future, I think labels such as “3D artist,” “AI artist,” or even “designer,” which are often defined by the tools someone uses, may become less meaningful.

How do you feel about AI? I saw some AI work you experimented with in 2023. Do you still work with it?

I am very interested in AI.

When I first experimented with image-generation AI in 2023, it was already exciting, but the technology has developed so much since then that it is difficult to compare the two periods.

I don’t really see AI as a tool for producing a specific visual style. I see it more as another method of making and exploring images.

I still use it for things like exploring references and testing ideas quickly, and I am increasingly interested in how AI can be integrated more naturally with 3D workflows.

I don’t think AI simply makes existing technical skills irrelevant. If anything, as the cost of creating images decreases, the ability to judge which images are worth pursuing and how they should be developed becomes even more important.

That is the part I find most interesting.

What’s one lesson from your transition into art direction that you think other 3D artists could apply to their own work right now?

Spend a little more time thinking about what you want to make before thinking about how to make it.

3D artists are generally very good at solving technical problems, so when we see a new project, it is easy to immediately start thinking about the production method.

I used to do the same.

But if you first define the purpose of the image, the emotion it should create, or even the single most important frame in the project, the final result can become very different.

A technically difficult piece is not necessarily a strong piece. Sometimes a very simple execution with a clear direction can be much more powerful.

If 3D artists want to explore the possibilities of art direction, where would you recommend them to start?

I would recommend directing a personal project from beginning to end.

Executing a brief that has already been defined is very different from starting with nothing and deciding the direction yourself.

Choose a subject, gather references, decide what belongs in the project, decide what does not, and try to build a small but coherent world.

It does not need to be technically ambitious.

Even three images or a ten-second film can be a very good exercise if you can clearly explain why you chose that color, that camera, that movement, and that rhythm.

I think that kind of practice is one of the best ways to begin developing art direction skills.

What are you personally most curious to explore next in your work?

Right now, I am most interested in removing the boundaries between different production methods.

I am curious about how procedural and physically based systems such as Houdini, which I have worked with for a long time, can be combined with something as fast and unpredictable as AI within a single creative process.

I am also becoming increasingly interested in work that extends beyond the screen into physical spaces, installations, and interactive media, where the audience can experience the work spatially.

Ultimately, I am less interested in becoming better at one specific tool and more interested in using new technologies to create images and experiences that I have not seen before.