ILM's Senior Lighting TD On VFX Challenges, Building Hologram Lighting For Transformers One & Entering The Industry

Could you introduce yourself, summarize your career, and share how your background growing up between Canada and Hong Kong shaped your eye as an artist?

My name is Jonathan Wai. I'm a Senior Lighting Technical Director at Industrial Light & Magic in Vancouver, with about 14 years in the industry. I was born in Canada but grew up mostly in Hong Kong, and eventually moved back to Winnipeg before relocating to Vancouver for film school.

Those two places gave me pretty different visual references. Vancouver is all natural light, overcast skies, and forests. Hong Kong is the complete opposite – neon everywhere, dense urban environments, light bouncing off glass, and wet pavement at every angle. I didn't fully realize how much that shaped my eye until I was deep into production work and kept finding myself drawn to how artificial light interacts with surfaces. That's still the thing I find most interesting about this job. 

What first led you into lighting for animation and VFX, and what were the biggest turning points in your 14-year career?

I've always had a passion for art, animation, and film from a young age. I studied art at Sheridan College and then transitioned to Vancouver Film School, which is where I was first introduced to lighting. It felt like painting all over again, except instead of actual paint, I was using light to tell a story and create a mood in a scene. That's when I decided to specialize as a Lighting Artist. From there, I worked my way through a few different companies and eventually grew into a lead role.

The biggest turning point was joining ILM. I came in at a junior level even though I had lead experience elsewhere, and that was a deliberate choice. I knew the pipeline and quality of work there was on a completely different level, and I wanted to learn it properly and keep growing as an artist.

That decision paid off. Working on The Mandalorian early on forced me to understand live-action integration at a depth I wouldn't have gotten anywhere else. The other major turning point was Spider-Verse at Sony Pictures Imageworks. That production had no rulebook. We were inventing the look as we went, and having to problem-solve at that level completely changed how I think about lighting as a craft rather than just a technical task.

For readers less familiar with the role, how would you explain what a Lighting TD does on a major feature or VFX production?

The simplest way I describe it is that a Lighting TD sits between the artistic vision and the technical execution. You are responsible for making the shot look the way the director and VFX supervisor want it to look, but you are also responsible for building the systems that make that possible across an entire sequence, not just one shot.

On a big production, you are designing light rigs that other artists will use, making sure things render consistently across hundreds of shots with minimal adjustments needed, and solving technical problems that come up when the original plan meets reality on screen. It is equal parts artistic and engineering, which is what I love about it.

You've worked on projects like The Mandalorian, Spider-Man: Into the Spider-Verse, Transformers One, and Lilo & Stitch. How have those different productions challenged you as a Lighting Artist? 

Each one was a completely different problem. Spider-Verse was about breaking rules. The whole point was that it shouldn't look like anything that had been done before, so you couldn't rely on established workflows. The Mandalorian was about volume and speed, hundreds of shots with live-action plates, StageCraft LED environments, and very tight schedules. Transformers One was an animated feature with its own visual language, but ILM's pipeline is built for photoreal work, so there was a constant calibration happening between the stylized look the directors wanted and what the tools naturally want to produce.

Lilo & Stitch was a different kind of challenge altogether. Stitch originally comes from a 2D cartoon, so there is this constant balancing act between staying true to how he looked and felt in the original animation while still making him feel grounded and believable in a live-action world. You want audiences who grew up with that film to recognize him immediately, but he also has to hold up against real photography and real environments.

One of the sequences I worked on involved a disco lighting effect that didn't have an obvious existing solution, so I ended up building a procedural system from scratch using animated light textures to drive multiple lights parented to the disco ball geometry. The goal was to make it physically accurate but also reusable, so other artists could apply the same setup across multiple shots without starting from scratch. I also developed a look development macro for the show that allowed artists to switch between approved surface variations with a single control, helping keep the look consistent across a large number of shots and adopted across the beach sequence. Both of those came out of necessity, but that is usually how the most useful tools get made.

What was the transition like moving from animation lighting into live-action VFX at ILM, and how did it change your approach? 

It was a significant shift. In animation, you have complete control. Every element in the frame was created digitally, so if something doesn't work, you can change it or even cheat the lighting to achieve the look you want. You have that creative freedom to be reactive. In live-action VFX, you are always working in the service of photography that already exists. The light in the scene was captured on set months ago, and your job is to make the CG elements feel like they were always there. That changes how you think about lighting fundamentally. You are not designing from scratch, you are matching, extending, and integrating.

In terms of how it changed my approach, it made me much more disciplined about studying real light. When you are matching a plate, you cannot fake it. You have to understand why light behaves the way it does physically, how it wraps around surfaces, how it falls off across distance, and how different materials respond to the same source differently. That level of observation has fed back into my animation work as well. I think understanding real light makes you a better Lighting Artist across the board, regardless of what type of production you are working on.

What are the biggest differences between lighting for animation and lighting for live-action VFX?

Control versus constraint is the short answer. In animation, unless you're given concept art from the director, you light from scratch every time. You decide where the sun is, what the ambient looks like. In live-action VFX, the plate dictates the environment, and your job is to honor it while making your CG elements feel native to that world. The technical challenges are also different. In animation, you are usually optimizing for consistency and style. In live-action, you are fighting for physical accuracy because the light behavior has to match real-world physics, or the eye catches it immediately. That said, the skills cross over more than people expect. Understanding physical light behavior makes you better at both.

How did Hong Kong's neon city atmosphere influence your lighting work on Transformers One, especially the skyscraper and city lighting? 

This one felt personal. The neon city sequences in Transformers One, the skyscrapers, the underground environments, they are visually set on Cybertron, but they read like a science fiction version of Hong Kong. Growing up there, you absorb a lot without realizing it. The way neon bleeds color onto surrounding surfaces, how light wraps around geometry in a dense urban environment, and the relationship between darkness and pools of colored light. That is not something you learn from a textbook.

When I was building the lighting for those sequences, I kept referencing that visual memory and the references I had collected over the years, how the city feels at night from street level, how light behaves on reflective surfaces in the rain. I think it gave those sequences a quality that feels grounded even though they are completely fantastical.

Could you walk us through the projection-based hologram lighting workflow you developed for Transformers One? 

The challenge with holographic light sources in CG is that the standard approach, using an area or rectangle light with an image plugged in as a light texture, comes with a lot of limitations and often requires significant fixes from the compositor. What I developed instead was a system where the hologram's visual content, the moving image, the flickers, the color shifts, and the animation were projected onto a sphere geometry using image sequences passed to me from the concept artist. That sphere then became the actual light emitter in Katana.

So instead of a static light with animated values, you have a physical shape emitting light based on real moving imagery. The result is that the light falloff, the color bleeding onto nearby surfaces, and the shadow behavior all of it responds correctly because the system is physically grounded. It is the difference between relying on multiple rectangle lights that could double or triple the light value and cause overexposure, versus having a single accurate hologram light source in the scene. In addition, the system is driven entirely by the image sequences fed into it. The same rig architecture can produce fire flicker, neon pulse, or strobe lighting just by swapping the imagery,  making it useful well beyond hologram effects.

Why was it important for the hologram projection to function as a real light emitter in the scene rather than only as a post-production effect?

Post-production fixes can become time-consuming and often require additional manual work. If a hologram is added as a comp element, you lose all the physical interaction. There is no color bleed on nearby characters, no accurate shadow behavior, and no correct falloff on surrounding geometry. Compositors then have to manually paint in all of that interaction frame by frame,  which rarely holds up under close inspection and is very time-consuming to fix in comp.

In Transformers One, specifically, almost all the characters are robots with highly reflective metallic bodies. By having the hologram function as a real physical light source, you can actually see the hologram image reflected in the characters' surfaces, which gives the scene a level of realism that would be nearly impossible to achieve in post. On top of that, a static light is limited by its placement and the direction it is facing. Having a physical shape emitting light means the illumination covers 360 degrees around the hologram and responds naturally to any flicker or movement in the source imagery. The shot is more believable, and the downstream compositing work is significantly reduced. Building things correctly from the start always produces better results than patching them in post, and it is more efficient in the long run.

How did you balance artistic quality, shot consistency, and render efficiency when building that reusable lighting setup? 

These three things are always in tension, and the balance looks different on every show. For the hologram system specifically, the key decision was building it to be sequence-level rather than shot-level from the beginning. I set up a default camera to act as a projection source pointing straight at the sphere, so all another artist needs to do is plug in the setup and swap out the image sequences for their specific shot. You will get consistent lighting results regardless of which shot you are working on, because the system is driven entirely by what image sequences you feed into it. That upfront investment pays off across the whole sequence, because the setup was reusable and sequence-based, other artists were able to adopt it without rebuilding the lighting from scratch, helping maintain consistency across multiple shots while reducing setup time.

On the render efficiency side, I was careful about how many lights were in the rig, how the textures were structured, and removing any extra assets that fell outside the camera threshold without affecting the overall look too much. You can build something that looks amazing but completely kills your render times, and that is not a real solution in production. The goal is always to find the approach that is physically accurate, visually correct, and actually deliverable within the schedule.

What tools, software, or pipeline considerations were most important to your lighting work on Transformers One?

The core of my work at ILM is Katana for lighting and look development, with RenderMan as the renderer. For the hologram system specifically, I used Nuke to process the image sequences handed off from the concept artist. I would add additional glow and flicker, or reduce any overexposed values to even out the lighting, before feeding the sequences into Katana and using them as the source for the projection lighting. The most important pipeline consideration was making sure the system could work reliably across the show, and potentially across other ILM productions if other teams wanted to implement it in their own pipeline. That meant keeping naming conventions and file paths consistent, and using variables rather than hard-coded references so the setup would work across different shots and shows without breaking. It sounds like a small thing until a production goes wrong because of it.

What advice would you give to artists trying to break into VFX lighting or  grow into more senior creative roles?

Two things. First, build tools that other people can use. The artists who grow fastest are the ones who think beyond their own shots and ask themselves how what they are building could help the person sitting next to them. That mindset is what gets you noticed and trusted with more responsibility. Second, be willing to step back to move forward. I took a junior title at ILM after being a lead elsewhere because I knew the environment would make me significantly better. That was the right call. Don't let a title stop you from putting yourself in a situation where you will learn more than anywhere else. The title catches up eventually.

Outside of the work itself, getting involved in the broader industry community has been something I wish I had prioritized earlier. I am a member of the Visual Effects Society and the Television Academy, where I am part of the Special Visual Effects peer group and eligible to vote in the Emmy Awards. I also serve as a judge for The Rookies, which is one of the largest international competitions for emerging VFX and animation talent. Those connections have given me a much wider perspective on where the industry is going and what skills are actually valued at the highest level. If you are serious about a long career in this field, showing up beyond your day job matters more than most people realize.