I did not plan to make any recap for RenderCon 2026. I thought it would be more or less the same as before. It was not.
This year I was one of the speakers, and I did not do public speaking since Video Copilot tour in 2019. I had 30 minutes to present The Pipeline. Very quickly I realised there is no honest way to explain all the intricacies in such timeframe, so several days before the talk I changed the whole presentation. Risky decision, but correct one.
Instead of trying to overload people with technicalities, I told the real story of how this pipeline happened. From selling Out of Nothing Shares and giving up old workflows, to building a CPU render farm, to landing larger commercial projects, to meeting production problems that only appear when things get serious in scale. Clarisse was a big part of that road too. All of this eventually led me to the concise version of The Pipeline with Octane.
And I think this was the right move, because I heard a few reactions like, “wow, I didn’t realise this.” That made me happy. I am on a mission with this work. I want to spread this craft and this way of thinking further, because I want this pipeline to become almighty, convenient and accessible not only for big studios, but for everyone. Studio or lone wolf.
Each person I met there would confirm the same thing: RenderCon is not like other conferences. For Octane community, it is unbeatable density of like-minded people. Not only same industry. Same tools, same pain, same ambition, same curiosity about where all this is going.
What made RenderCon 2026 especially interesting to me is that, beneath all the noise around AI, the event was actually very close to real CG and VFX production questions. Creative intent. Real deadlines. Cinematic pipelines. Control. Authorship. How new technology can enter production without destroying the artist’s hand. This is the part that interests me most.
For example, in my talk I explained how The Pipeline already gives production superpowers now, while also preparing artists for future workflows around Neural Objects and similar neural representations. We also saw Jules talking in this direction again. He has been pointing there for years, and this is exactly why I tend to take those predictions seriously. When someone is early for long enough, eventually you understand they are not guessing.
One thing becomes more and more clear to me: the future of commercial-scale production is not in generative AI alone. It is in controlled pipelines that can absorb new technology without sacrificing structure, quality and authorship. And this fact does not make me anxious. It makes me curious.
I also really enjoyed Refik Anadol’s talk. Seeing projects around Dataland and Large Nature Model was a strong reminder that there is a huge difference between AI as a toy and AI as part of a larger artistic or production vision. The exciting part is not pressing a button. The exciting part is imagining a future where neural assets can become real, controllable building blocks inside a high-end pipeline.
This is why I hope I will be back at RenderCon next year. Everyone I spoke to said the same thing: it is one of a kind. I am grateful I had a chance to be part of it.
I did not plan to make any recap for RenderCon 2026. I thought it would be more or less the same as before. It was not.
This year I was one of the speakers, and I did not do public speaking since Video Copilot tour in 2019. I had 30 minutes to present The Pipeline. Very quickly I realised there is no honest way to explain all the intricacies in such timeframe, so several days before the talk I changed the whole presentation. Risky decision, but correct one.
Instead of trying to overload people with technicalities, I told the real story of how this pipeline happened. From selling Out of Nothing Shares and giving up old workflows, to building a CPU render farm, to landing larger commercial projects, to meeting production problems that only appear when things get serious in scale. Clarisse was a big part of that road too. All of this eventually led me to the concise version of The Pipeline with Octane.
And I think this was the right move, because I heard a few reactions like, “wow, I didn’t realise this.” That made me happy. I am on a mission with this work. I want to spread this craft and this way of thinking further, because I want this pipeline to become almighty, convenient and accessible not only for big studios, but for everyone. Studio or lone wolf.
Each person I met there would confirm the same thing: RenderCon is not like other conferences. For Octane community, it is unbeatable density of like-minded people. Not only same industry. Same tools, same pain, same ambition, same curiosity about where all this is going.
What made RenderCon 2026 especially interesting to me is that, beneath all the noise around AI, the event was actually very close to real CG and VFX production questions. Creative intent. Real deadlines. Cinematic pipelines. Control. Authorship. How new technology can enter production without destroying the artist’s hand. This is the part that interests me most.
For example, in my talk I explained how The Pipeline already gives production superpowers now, while also preparing artists for future workflows around Neural Objects and similar neural representations. We also saw Jules talking in this direction again. He has been pointing there for years, and this is exactly why I tend to take those predictions seriously. When someone is early for long enough, eventually you understand they are not guessing.
One thing becomes more and more clear to me: the future of commercial-scale production is not in generative AI alone. It is in controlled pipelines that can absorb new technology without sacrificing structure, quality and authorship. And this fact does not make me anxious. It makes me curious.
I also really enjoyed Refik Anadol’s talk. Seeing projects around Dataland and Large Nature Model was a strong reminder that there is a huge difference between AI as a toy and AI as part of a larger artistic or production vision. The exciting part is not pressing a button. The exciting part is imagining a future where neural assets can become real, controllable building blocks inside a high-end pipeline.
This is why I hope I will be back at RenderCon next year. Everyone I spoke to said the same thing: it is one of a kind. I am grateful I had a chance to be part of it.
