The portfolio — Building it backwards
I haven't built my portfolio yet. This is me figuring out how to.
When I first started thinking about a portfolio, the logic seemed obvious: make good work, put it somewhere people can see it. But after looking into what studios actually want, I realised that's the wrong starting point. A portfolio isn't just a collection of things you've made, it's also the impression you leave on someone who has never met you. That impression comes down to what you put in, what you leave out, and what someone sees first. So it makes more sense to start from the conclusion. What do I want someone to think about me, and work backwards from there.
For me it's: this person can take a brief, whether that's a full project or just one part of the pipeline, and bring it to a place where you can see the thinking behind how they got there. In the industry you often own just one piece of a much larger production. A modeller doesn't deliver the final render. An FX artist doesn't do the comp. The portfolio doesn't need to prove you can do everything. It needs to prove you understand your part, can execute it to a professional standard, and can show your reasoning clearly enough that someone else could pick it up from there.
The three tiers
Not everything you make belongs in the same place.
The showreel is the elevator pitch — sixty to ninety seconds, best work only. It isn't a highlights reel of everything you've made. It's a case you're making to a specific person in a limited amount of time, and every shot needs to earn its place. My favourite quote from this research was from a TD with credits at Wētā FX and Framestore: "Best work first, second best last and nothing in between that brings it down. One weak shot doesn't just sit there, it brings your level down."
ArtStation is the public feed. Think of it like Instagram, you post work you're genuinely proud of, and inside each post you go into more detail about the process. Everything visible at the top level needs to hold up on its own.
The website is everything. More process, more context, work that doesn't make the showreel but shows range or a particular decision you made. It's also what I'd send if someone asked for my portfolio directly. It gives the most control over what they see and in what order.
What the showreel actually needs to show
Studios aren't just evaluating whether your work looks good. They're trying to answer a more specific question: can this person function inside a production?
The breakdown is where that gets answered. Not a technical deep-dive, but enough to make your contribution clear. What the shot is, what you specifically did, which software you used, and what the before vs after looked like. A recruiter who has reviewed hundreds of reels isn't going to give you the benefit of the doubt if it isn't obvious what you did.
For FX work, showing the simulation process and the real-world reference you were matching is worth including. The goal is to show you understand why something looks the way it does, not just that you ran the sim.
My actual situation
The work I'm building toward a portfolio with right now is a gladiator character — sculpt, cleanup, texture, props, lookdev, render. A separate Houdini cloth simulation. And two Nuke comp shots.
The gladiator is finished in the sense that I drew a line on it. I got to a point where fixing certain things would have meant going back significantly, and the deadline didn't allow for that. So I stopped. The showreel version will need a proper breakdown. The turntable and close-up renders exist, but showing the pipeline, sculpt passes, texture sheets, and lookdev decisions is what turns it from a nice render into an argument for making decisions to create a finished outcome.
The Houdini cloth piece was a trouser tear for a creature attack shot. The brief was to make it look powerful. I went through several different methods, pre-fracture, noise-driven tear thresholds, before combining approaches and iterating until it was working the way it needed to. That kind of iteration, trying something, understanding why it isn't right, and adjusting, is what I'd want the breakdown to show, not just the final result.
The Nuke comp work is interesting too. I had two shots with broken plates — one with no cryptomatte or Z-depth pass, exposure issues, and a brief to make it feel like footage from a real 2002 Sony camera. The other shot also needed to control exactly when the viewer could feel the bear's presence — barely visible at first, just a nose outline and a strange shape in the lake, then fully revealed during a lightning strike. I covered animation errors using camera scanline artefacts and controlled the colour and tension of both scenes. The shots were close to being cut. They weren't. That's what a breakdown of those shots would need to communicate — not just the technical fixes, but the decisions behind them that saved it from being cut.
The standard
The clearest test I've found: Would you be comfortable showing this to a client? Not a beauty render necessarily — a playblast is fine if the animation is genuinely done, a viewport screenshot is fine if the model holds up. The question is whether the core work is there and whether it demonstrates the specific skills the piece is supposed to show.
Effort and professional readiness aren't the same thing. Some of the work I've spent the most time on probably shouldn't be on my ArtStation — not because it's bad, but because it isn't ready for that context. Figuring out which is which is something I'm still working on. I did a work experience placement at ETC and met people there I'm still in contact with — one mid-level CG generalist is doing a video call with me to give feedback on my work and answer questions. That outside perspective, from someone actually working, feels more useful than any internal checklist.
Building the portfolio is the easier part. The harder question is whether the skills inside it will still be what studios are looking for by the time it matters.