Introduction

Every few weeks there’s a new AI tool that promises to do someone’s job at the click of a button. Some creative jobs really are changing fast. So I wanted to talk to someone whose work looks, at least from the outside, like an obvious target for automation: a colorist.

Nicole Goode is a professional freelance colorist. She has graded feature films from Hollywood to Bollywood, hard-hitting documentaries, family films, adverts for big brands and student short films. Color grading is a very specific job, and few people outside the film industry know what it involves. That made me even more curious about how AI affects it.

We talked about what a colorist actually does, how she finds clients as a freelancer, and how to start a career in color grading. We also talked about why, surprisingly, AI is almost useless in her work. This article summarises the key points from our conversation.

What a colorist actually does

Footage doesn’t come out of the camera looking like a film. Shots from different days, cameras and lighting conditions all look slightly different. The colorist takes all of that footage, balances it so that it looks consistent, and then gives it the final look. Many people on a film set work towards the same goal of a beautiful film. In Nicole’s words, the colorist is the person who adds the final icing on the cake.

A colorist’s job has an unusual rule: if the audience notices your work, something has probably gone wrong. Cinematographers and other film people might watch a film and admire the grade. Most viewers are there for the story, and anything that doesn’t fit pulls them out of it. The same is true of sound, cinematography and every other part of filmmaking.

People shouldn’t be thinking about my work when they’re watching a story.

So the goal isn’t to push the image as far as possible. It’s to find what works for the story, so that the audience accepts the world on screen and simply experiences the film.

Life in a dark room

Much of the job means sitting alone in a blacked-out room, turning knobs. The only light comes from bias lighting, a dim, cool light behind the screen that keeps the atmosphere consistent. It reduces eye strain and gives your eyes a stable reference point. Without it, your brain can start to misjudge what you see.

Nicole works eight to twelve hours a day, sometimes seven days in a row, without seeing the sun. Once she’s absorbed in the work she doesn’t notice, but eventually it takes its toll. Her first piece of advice for future colorists was, half-jokingly, to take vitamin D. And when you’re grading a scene set on a beautiful sunny day, you can always pretend.

Clients come from people who trust you

How do you find clients as a freelance colorist? Nicole’s answer is honest: there’s no secret, other than slowly building a base of people who know and trust you. It’s the unexciting answer, but it’s the one that works.

The way to do it is to play to your strengths. After graduating from film school, Nicole worked in the school’s editing department and taught workflow to students. She met many young filmmakers there. When they later went on to make their own films and she started her freelance color grading business, they remembered her.

At first, that meant lower-budget work and paid student films. It wasn’t big money, but it was enough to build a showreel and a client base.

Be visible in as many places as possible

Word of mouth isn’t Nicole’s only channel. Her website has brought in clients who simply searched for something like “colorist in Europe” or “colorist Prague”. She also keeps an Instagram account for work purposes. She’s not very active there, but the account exists.

The idea is to keep popping up wherever potential clients might look: in Google search results, on Instagram, on Facebook and in recommendations from people they know. When someone sees you in several places, you look like a real, established professional, and hiring you feels like a safe choice.

A little self-promotion here: Nicole’s website is built with Minimalio, my WordPress theme made specifically for creators who want to present their work. If you’d like a portfolio website like hers, there’s an all-in-one tutorial on this website that takes you through everything step by step.

AI is a non-factor

I expected color grading to sit somewhere in the middle. On one end are jobs where AI can do most of the work at the click of a button. On the other are jobs that barely involve a computer at all. Nicole’s answer surprised me.

In short, for me, it doesn’t. It’s kind of a non-factor in my day-to-day life.

Many people see AI as a shiny new toy that will make their workload lighter. For a colorist, though, it has very little practical use at the moment. I expected it to help with at least some of the routine work, such as matching shots to each other. From what Nicole has seen, it can’t even do that well.

Color is perception, not just code

Why does AI struggle with grading? An image is just data, and so is color. But a huge part of how we see an image happens in our own eyes, and our eyes read colors differently depending on what surrounds them.

Nicole gave a simple example. Put a big yellow billboard behind her and her skin will look sallow and slightly green. Put a color that suits her better in the background and she’ll look completely different. The pixels of her face haven’t changed, but our perception of them has.

A computer doesn’t know the difference between my skin tone against yellow or blue, but our eyes see a huge difference.

This is called the simultaneous contrast effect, and it’s a big part of the invisible, technical side of grading. Accounting for all of these small perceptual shifts is very hard to do with an algorithm. Nicole doesn’t see this as a weakness of the job. She sees it as a strength: a trained human eye is exactly what the work needs.

A trained eye is faster than AI

There’s also the question of speed. An experienced colorist grades a shot in less than a minute, and on a busy day Nicole can work through around 700 shots. Modern camera files are huge, so running AI over a whole timeline of a few hundred shots would take an enormous amount of processing power. It isn’t practical.

The AI grading examples Nicole has seen online haven’t impressed her either.

You’re working too hard trying to use AI. I could have done that in a minute.

For now, AI in color grading is a longer route to worse results. Nicole isn’t opposed to AI in general, and she’s keeping a neutral view to see where it goes. But she doesn’t have high hopes, and she isn’t eager for it to take over. She enjoys the flow of grading shot by shot, and she wouldn’t want to lose that.

Is color grading the right career for you?

The job suits people who enjoy creating good-looking images, like working on a computer and are happy working independently. But it isn’t a solitary job. Even though you spend much of your time alone in the dark, a large part of the work is with clients. That’s why Nicole thinks the most important quality is being a nice, patient person.

Learn from the manual, not random tutorials

The first step is to find out whether you enjoy the work, then start practising and training your eye. The challenge is that there are so many resources out there. Search YouTube for “how to color grade” and you’ll find great tutorials alongside terrible ones, with no easy way to tell which is which. That’s how beginners pick up bad habits.

Nicole’s recommendation is refreshingly simple:

Just read the DaVinci Resolve manual. It’s a good read. It’s like 4,300 pages, but you’ll learn a lot.

Start on a laptop, deliver on calibrated equipment

You don’t need a professional grading suite to start learning. You can experiment with DaVinci Resolve on a laptop. You shouldn’t deliver work to a paying client that way, though.

Professional work has to be graded on calibrated, industry-standard equipment. Many monitors can now calibrate themselves, and Nicole thinks EIZO monitors are probably the best place to start. They’re expensive, but that’s the reality of the job. Once you have a proper monitor and a dark room, you can gradually build out the rest of your suite.

Leave your ego at the door

Nicole’s final piece of advice was about people rather than technique. Early in her career, she got defensive when clients didn’t like something. Over time, she learned to accept their feedback and move on. If a client doesn’t like it, they don’t like it. Sometimes the problem isn’t even the grade: the client might be watching on a badly set up screen, or with night mode turned on.

To set the right tone from the start, Nicole begins every project with a look development session. It hands control back to the client. After all, the film is their baby, not hers.

If you want to be a colorist, don’t make it about you. Make it about your client and their baby.

Key takeaways

• Great grading is invisible. Serve the story, not your showreel.
• Build trust over time. Most clients come from people who already know your work.
• Be visible everywhere. A website, Instagram and a presence in Google search all add up.
• AI isn’t a threat to colorists yet. Color depends on human perception, and a trained eye is faster.
• Read the manual. The DaVinci Resolve manual beats random tutorials.
• Invest in calibration. Learn on a laptop, but deliver on calibrated equipment.
• Drop the ego. Accept feedback and make it about the client’s vision.

Many thanks to Nicole for such an honest and fun conversation. If you’re curious about color grading or wondering how AI will affect creative work, watch the full interview for the complete conversation.