INTRODUCTION

For the last few years, my world has been VFX compositing, cleanup, roto, FX, and everything that goes into building the final picture. I’ve worked with traditional pipelines, industry tools, long renders, and the usual deadline chaos.
But recently, AI has started to shift the industry in ways we can’t ignore. Instead of fighting it, I decided to understand it deeply and maybe turn it into an advantage.

This blog is where I’m documenting that journey.

1. WHY I’M EXPLORING AI NOW

AI isn’t replacing real VFX work, it’s changing how we work.
Tools like ComfyUI, SDXL, and image-to-image workflows aren’t just shortcuts. They offer new ways to prototype, visualize, and build ideas faster than traditional pipelines allow.

For me, experimenting with AI isn’t about trends. It’s about:

• Speeding up small parts of the pipeline

• Exploring creative concepts

• Reducing repetitive tasks

• Boosting my ability to work independently

I want to take what I already know as a compositor and expand it into a hybrid workflow.

2. WHAT THIS BLOG IS ABOUT

I’m building this blog to share:

• AI workflows I’m learning

• Experiments inside ComfyUI

• How I combine AI with real VFX techniques

• My small short-film tests

• Practical, no-bs breakdowns anyone can follow

Every post will be structured, clean, and easy to repeat in a real pipeline.

3. FIRST GOAL BUILDING CONSISTENT WORKFLOWS IN COMFYUI

Instead of using random prompts and one-click tricks, I’m focusing on building repeatable workflows.

Things I’m currently testing:

• Object removal

• AI-assisted rotoscoping

• Cinematic frames

• Inpaint-based cleanup

• Custom character creation

• Small shot sequences for short films

My aim is simple:
To build workflows that feel like real VFX processes but faster.

4. HOW I’M APPROACHING THIS JOURNEY

I’m treating AI like another tool in the VFX toolbox, not a replacement.

My approach is:

• Experiment

• Document

• Refine

• Share

This blog will cover both successes and failures. If something breaks, I’ll explain why. If something works, I’ll break down the exact steps.

5. WHAT’S COMING NEXT

Over the next few weeks, I’ll post detailed guides such as:

• Cinematic Shots in ComfyUI (Step-by-Step).

• Object Removal Workflow (Production-Style).

• How I Build AI Concepts Into Real Shots.”

• My Short Film Experiment: The Last Dream.

This is just the starting point.

6. RESOURCES

Here are some useful tools, communities, and references I’m using:

• ComfyUI — Main AI workflow software I’m experimenting with.

• Stable Diffusion XL — For high-quality image generation and inpainting.

• Lexica — Great for prompt inspiration and reference images.

• Hugging Face — Repository for AI models, including inpainting and text-to-image models.

• Nuke — Industry-standard compositing tools to combine AI outputs with real VFX pipelines.

These resources are where I pull references, models, and workflow ideas for every experiment.

CONCLUSION

This blog is my space to explore, test, and build.
Not everything will be perfect, but everything will be real and honest.

If you're into VFX, AI workflows, or want to follow the experiments, welcome.
This is going to be interesting.

Follow Me

Let's build something great together