Stéphanie Roy – 3D Supervisor, Cinesite
Stéphanie is a CG Supervisor at Cinesite, some of her recent projects include: Animal Farm (2025), Iwájú, Paws of Fury, Extinct, Fearless, and Riverdance: The Animated Adventure.
Stéphanie will be part of World VFX Day this December, talking about how she got involved in AI, the basics of using AI and Machine Learning, and how it can be applied to VFX.
Taking it back to the beginning where you studied Art and Science of Animation at University, did you have a clear idea of the type of career you wanted from an early age? What was your inspiration?
I grew up in a household where films were at the centre of everything. My parents were both cinephiles – we had an entire wall covered with VHS tapes. My dad, especially fond of horror films, secretly introduced me to the genre when I was five. Those hidden viewings sparked my love for horror and gothic worlds, especially Tim Burton’s work. By twelve, I wanted to be a horror movie director, so my parents helped me prepare: I took acting lessons to understand directing, attended an English-speaking school for a year to broaden opportunities and took drawing lessons to bring my visions to life. In a pre-university film program, I discovered many directors I admired started in visual effects, which led me to study Computer Graphics and Animation.
I still hope to direct a horror movie one day—or at least work on one—but for now, the Animation/VFX industry has brought so many challenges and so much learning that I truly feel I made the right decision.
Cinesite’s VFX Showreel:
Who have been, or are, your biggest career inspirations or role models so far?
At the beginning of my career, I often found myself without leads or supervisors to look up to. I learned to draw inspiration from my peers and from individuals outside of our field. I wouldn’t say I have a single role model; instead, I try to identify what makes leaders great and work on developing those qualities within myself.
You’ve been at Cinesite for almost 10 years. Did you have a career plan, or did it happen more organically? Tell us about your journey at Cinesite so far.
In university, I specialised in rigging because it was in-demand, technical and clear-cut, unlike more artistic fields. After three years as a rigger at MELS and Framestore, I joined Cinesite, where my leadership skills grew, eventually leading the rigging team. After working on a few movies, I realised I wanted to be involved in more than just this one department.
“I wanted a say in decisions made across the production and to have a better understanding of the entire movie-making process.”
I shared this with Cinesite who helped me with a transition-plan to become CG Supervisor.
The initial plan was for me to shadow another CG Supervisor on one movie, with only one sequence to handle independently. But when an additional movie came up unexpectedly, that CG Supervisor was reassigned, and I was offered the chance to supervise a full movie on my own. It was incredibly challenging, but the Heads of Departments and all team members were incredibly supportive and welcoming. I ended up absolutely falling in love with the role.
Which projects are you most proud of, or most enjoyed, and why?
My two favourite projects were Blazing Samurai and Iwaju. Blazing Samurai was my first movie as a solo CG Supervisor, where I felt truly confident in the role and had the opportunity to test new technologies and push the team forward. It was a lot of fun to be a part of! Iwaju was our first partnership with Walt Disney Animation Studio, and I’m incredibly proud of the result. It was our highest-budget project to date and Disney’s quality expectations were much higher than what we usually delivered. I watched my team grow rapidly and I’m extremely proud of how much we learned from the experience.
Left: Blazing Samurai, Right: Iwájú
What were some of the biggest challenges you encountered early in your career and how did you overcome them?
Not feeling like an impostor! Because I was very young when I became a CG Supervisor and only had experience with the front end of a production, I often worried about letting people down if I didn’t know the answer or the right direction to take.
“I’ve learned that it takes much more courage to say, “I don’t know, let me find the answer and get back to you,” or, “I tried, and it didn’t work; let’s try something else!” than to stay silent or pretend you know everything just to avoid looking incompetent.”
We work in a constantly evolving field with new technologies emerging every day—it’s normal not to have all the answers. The important thing is to keep asking questions, keep learning, and not fear failure. Being afraid to fail leads nowhere, as it will only hold you back from trying anything new.
You have been delving into the world of AI and how it can be used in VFX. How are you using it and what have been the biggest benefits so far?
We’ve been exploring AI’s potential in VFX for a few years, starting with tools like machine learning for denoising and face replacement. With the recent influx of new ML tools, our focus has shifted to finding ways to empower our artists. The goal isn’t just automation but helping them reclaim time spent on repetitive or less creative tasks. This way, they can focus more on the work that excites and challenges them—the “fun stuff” that makes the job so rewarding.
We’re still early in our journey, having only started developing these new tools within the last year, but this guiding philosophy is clear: we use ML to support our artists, making them more efficient where it counts. It’s about enhancing their workflows, not replacing their talents.
What have been the biggest learning curves for you in terms of working with AI?
The biggest learning curve has been finding solutions that balance all the key elements we care about.
“There are new models coming out every day but few solutions are meeting our requirements from the moral, legal, security and quality perspectives.”
We’re focused on creating a flexible approach that respects each client’s vision and needs. Balancing the adaptability AI provides with robust security and privacy measures is complex but essential. This experience has reinforced our commitment to using AI responsibly, prioritising data privacy, intellectual property rights and a secure setup that benefits both our team and clients.
If someone wanted to learn more about how to ethically use AI in their work or studies, where should they start?
I would recommend looking into videos from Tanmay Bakshi or attending an online class. I started my journey with online classes from IBM (IBM AI Developer) and Lund University (AI, Business & the Future of Work).