shots Unsigned: Seth Ickerman

Behind the pseudonym of Seth Ickerman are two directors, Savitri Joly-Gonfard and Raphaël Hernandez. Their approach is built upon complete control of the creative process, they oversee every stage, from screenwriting and editing to visual effects, to bring to life independent works with a striking and original visual scale.

A significant part of their universe is defined by our musical collaboration with the artist Carpenter Brut. Their style is tinged with contemporary retro-futurism and mystical science fiction, drawing its roots from the legacy of Heavy Metal and the organic fusion of director David Cronenberg. This image-meets-music collaboration resulted in a trilogy with a powerful visual signature, a "Cosmic Opera" mythology told through the music video Turbo Killer, the film Blood Machines, and Speed or Perish.

By breaking free from the traditional constraints of the industry, we strive to bring ambitious and radical French science fiction to a global audience, where every frame is conceived as a painting and every project overflows with cinematic generosity.

Above: Some stills from Seth Ickerman's latest music video Speed of Perish for Carpenter Brut.

Can you tell us a little about your background and your route into directing?

Our journey is, above all, the story of a meeting between two cinema enthusiasts, Joly-Gonfard and Hernandez, who decided to merge their skills to create what they couldn't find anywhere else. Coming from backgrounds in drawing, stop-motion, and documentary filmmaking, our path to directing has been that of passionate self-taught creators, marked by a constant hands-on confrontation with the cinematic medium. Eschewing the traditional film school curriculum, we forged our identity through raw experimentation and a handcrafted approach to the craft.

Joly-Gonfard’s time at the Émile Cohl art school has also left a lasting mark on our vision: we don't see cinema as mere filming, but as a moving illustration where the visual and artistic form often takes precedence over academic codes. From our first experiments in the early 2000s, we laid the foundation for what would become our DNA: an ultra-lean team driven by immense visual ambition. We have constantly fed off each other’s respective skills, which, being so different, are perfectly complementary.

For six years, we locked ourselves away in near-seclusion in our "high-altitude studio" in the heart of the mountains. 

Our true baptism by fire came with Kaydara. For six years, we locked ourselves away in near-seclusion in our "high-altitude studio" in the heart of the mountains. This project was our real film school. By spending years on every single shot, we wanted to prove that with fierce determination, two creators could achieve a professional level of finish. This film was our "tour de force": it showed that we were no longer just technicians, but directors with a vision.

Following this early work, every subsequent project has continued to be our classroom, and that remains true today, as each new endeavour brings its own set of challenges and limits to push. Ultimately, our journey is defined by a commitment to the "long game" and a desire to prove that French-style sci-fi can be visually revolutionary, provided one accepts to turn constraints into a primary creative ally.

Above: Some production design on the latest film for Carpenter Brut.

Did you study filmmaking? How did you learn your craft?

Seth Ickerman is primarily self-taught. Our education didn't happen in a classroom, but through a constant and direct confrontation with the process of making films. It is through doing, making mistakes, and starting over that we forged our craft.

If we attended a school, it was for drawing, not for filmmaking. This curriculum deeply shaped our identity: we approach cinema with a gaze that is more artistic and plastic than purely technical or academic. For us, a film is often conceived as a moving illustration, where every shot is crafted with an aesthetic rigor inherited from the graphic arts.

We love to make everything ourselves. This "hands-on" approach allows us to maintain total control over the work, from the original idea to post-production.

Our journey stands at the crossroads of several fundamental pillars: We love to make everything ourselves. This "hands-on" approach allows us to maintain total control over the work, from the original idea to post-production.

Rather than suffering under budgetary or technical limits, we have learned to use them as a creative engine. Doing things differently is not just a choice; it is a necessity that defines our style.

Learning the trade as a self-taught creator also means accepting that the maturation of a project can take years. We work for the long haul, with the patience of an artisan, to ensure the final result remains faithful to our initial vision. In short, our "school" was one of raw experimentation, guided by a fierce determination to offer a singular cinema, liberated from traditional codes.

Above: Turbo Killer, a A hyper-stylised sci-fi synthwave music video is the first film of the trilogy.

Would you say you have a directing style? How did you arrive at it?

Seth Ickerman is a two-headed hydra with two very different personalities; this is a true creative asset that prevents us from stagnating in our comfort zones. Constraints are an integral part of the creative act, and our primary enemy is ourselves: self censorship and a lack of perspective. Our directing style is shaped by our constraints!

To create is often to create "against" something. Loving cinema also means disliking many things. You could say we make the films we would want to watch ourselves. Fighting against clichés and trying to unearth original ideas is where the true interest and passion of this profession lie. In our early days, we made a "thesis film" set in the Matrix universe (Kaydara), an interesting stylistic exercise, but one that appropriated the imagination of other creators. After that film, we were hungry to finally develop our own ideas and our own visual language.

It is difficult to self-define or dissect our own work

It is difficult to self-define or dissect our own work, but you could say we love the contrast between "tableau shots" a staging that takes its time, modest in camera movement, and a more "music video" dynamic, closely tied to the energy of musical rhythms. It is the contrast between contemplation and the sensation of high-velocity movement. All of this is intimately linked to the music, which we consider a character in its own right.

Of course, the style also depends on the projects and their specific universes, whether it is a short, medium, or feature-length film, not to mention the budget. Everything is interconnected in filmmaking, and you have to be pragmatic to see projects through to the end. The only moment you have total control is at the very beginning of the process; after that, the wave of constraints that hits forces you to readjust everything and make constant concessions.

To give a recent example: for the music video for GUNSHIP - Blood For The Blood God (Feat. HEALTH), we opted for long takes, the film has only 15 shots for seven minutes. Conversely, for Speed or Perish, there are about 200 shots for the same duration.

We can't say our absolute style is the long take or fast-paced "jump cut" editing; it simply felt relevant to take those directions given the nature of the project and the music. More than anything, we love testing new things and not repeating ourselves, which is far from simple.

When it comes to the visual aspect of our films, once again, we possess the aesthetics of our constraints. The trilogy in collaboration with Carpenter Brut clearly has a specific tone and style that we shaped in relation to the unique universe of his music. However, what we developed for those three projects does not define our style in a fixed or permanent way. Above all, we create according to the needs of each individual project.

Our role as directors is to maintain consistency and the global vision, holding everything together to make it solid.

In fact, we can’t say that we have a definitive style in mind at the start of a project. It’s actually the opposite: the style reveals itself at the very end of post-production, proving that the specific workflow of a project is what generates its visual signature.

We talk a lot about "constraints," but for us, they are the spice of creation. Creating without any constraints would be infinitely sad and an immediate source of limitation. We believe that the true definition of an artist doesn't lie in knowing exactly what they want or knowing the final result in advance, but rather in their ability to absorb and subvert constraints to their advantage. They allow for things to happen that we could never have imagined on our own.

It’s exactly like working with actors or any other collaborator on a project: everyone brings their own perspective and expertise, which makes the work evolve and keeps it relevant. Our role as directors is to maintain consistency and the global vision, holding everything together to make it solid. Because let's be real: the sheer volume of constraints during a film's production is so great that everything can quickly collapse and lose its shape. And if we were ever struck by a total lack of constraints, we would sabotage ourselves just to trigger some! For us, it is a matter of creative survival.

ABOVE: Speed or Perish, the most recent, and third instalment of the trilogy.

What is the inspiration behind your short films?

Our inspirations are never literal. Apart from Kaydara, we have never used a specific work as a blueprint or a starting point. Like all artists, we are shaped by the thousands of works we have loved; they leave a subtle imprint on the form our projects take. This process is unconscious, never intellectualized. We don’t make "tribute cinema" or films that are over-reliant on references.

However, we do have recurring themes, such as the body versus the machine, or the organic against the mechanical, how our civilisation grinds down the human body through its own inventions. The Wachowskis are never far off, nor are Cronenberg, Verhoeven, or Miller. This is a major trope in science fiction.

For Speed Or Perish, you could say we blended Gunnm (Yukito Kishiro’s manga) with George Miller’s Mad Max (especially the energy of the last two installments). Speed Or Perish was also a commission from Carpenter Brut, whose own influences draw heavily from 1980s sci-fi. The cynical, almost caricatured tone typical of that era was something very important to him. He wanted a cyberpunk car race where the last-place driver is summarily blown up to the sound of applause. We had to work with that.

ABOVE: A selection of BTS photographs from the making of Speed or Perish.

What other directors' work do you admire, and why?

Fortunately, our cinematic tastes are not confined to science fiction. Quite the opposite. What interests us most is seeing directors who offer a different vision; it allows us to break down our own barriers and find the courage to confront unknown territories.

Unfortunately, the industry loves to see patterns repeat and tends to go in circles. Artistically, it’s devastating. Luckily, dissenting visions occasionally emerge and offer something new. When they come from big names like Steven Spielberg, Denis Villeneuve, or Christopher Nolan, they are the ones who set the tone; they are then copied and re-copied for decades.

Making a list is always difficult, but here are our essentials: Terrence Malick, for his (meta)philosophical power tied to imagery and staging. Specifically The Thin Red Line and A Hidden Life. Paul Thomas Anderson, for characters that feel more real and raw than life itself. Specifically There Will Be Blood. Quentin Tarantino, because every single scene is captivating. You can tune into his films at any moment and be instantly hooked. He’s one of the few who understands that a film is more than just a script. George Miller, one of the rare directors who understands that a movie isn't just "dialogue by the mile." The older he gets, the more energy and talent he seems to gain (unlike many others, such as Ridley Scott, unfortunately). Paul Verhoeven, for his off-the-charts violence and his uncompromising, politically engaged themes. David Lynch, for infusing contemporary art into cinema and marrying humour with terror. Specifically Mulholland Drive. Andrei Tarkovsky, and Stanley Kubrick.

Lists are frustrating because they inevitably leave so much out.

In France, we have Jacques Audiard for his ability to consistently deliver masterful and unique cinema; Quentin Dupieux simply because he dares to do anything; and Gaspar Noé because he is the only one making a unique brand of cinema that hits you in the gut, quite literally.

Lists are frustrating because they inevitably leave so much out. We love our past references as much as discovering recent films. We try to balance the "untouchable" classics of the past so we don't become stagnant in our vision of cinema. Even if the industry "bulldozer" can be depressing, relevant works continue to emerge and still manage to move us.

Finally, cinema isn't everything: you have to admire a multitude of other things and step outside our own medium. We love the adage: "If you want to make a horror film, watch comedies." This applies to everything else. If you want to make action sci-fi, watch meditative social dramas! And if you want to make cinema, go to the theatre; or if you want to be creative, be bored.

ABOVE: A selection of BTS VFX images from Speed or Perish.

Where do you find the motivation for your projects?

We can distinguish between two types of motivation: the initial spark that provides the momentum to start, and long-term perseverance, since our projects always take several years to come to fruition!

The initial impulse that made us accept this project was the promise of AI. No sane person would agree to a cyberpunk death-race project, set in a futuristic city with 20 vehicles tearing each other apart, without colossal resources. The budget provided by Carpenter Brut for the music video was not insignificant, but given the desired ambition, it logically fell far short of what was actually needed. Producing this video through traditional means would have been unthinkable. But AI is shaking things up, suggesting (or rather, imposing) a complete rethink of our methods.

We had already experimented with it for the Blood for Blood God music video, but back in 2023, generating video via AI was still unthinkable. As technology evolves rapidly, we felt that something new was becoming possible for Carpenter Brut: a way to break free from, or at least reduce, the heavy burden imposed by the traditional VFX and CGI process. Reducing this constraint was supposed to allow us to focus on directing, strip away our self-censorship, and give ourselves "no-limit" creative freedom!

We are, first and foremost, directors. Even though we create our own images, what truly drives us is the staging and execution of our ideas.

Up until now, every film project we’ve undertaken has accounted for the complexity of special effects from the very start. We never went in blind, thinking anything was possible; we had to prepare and measure everything down to the millimeter to guarantee feasibility. Suddenly, with AI, a new way of doing things was emerging.

We are, first and foremost, directors. Even though we create our own images, what truly drives us is the staging and execution of our ideas. The promise of AI, producing the extraordinary at a lower cost, seduced us, and we wanted to challenge ourselves with this new method. When production constraints disappear, imagination takes flight. We wrote and conceived Speed Or Perish without any restraints. We focused solely on the writing and the ideas. It was almost exhilarating because, at our scale, it would normally be impossible to design a cyberpunk sci-fi film of this caliber without a Hollywood budget.

However, this beautiful promise of AI quickly turned out to be misleading and ultimately didn't help us as much as we had hoped. But the video was already in motion, and still in "no-limit" mode! We therefore reverted to the traditional way of filmmaking, which required a year and a half of relentless work. We would never have accepted the project had we known it would have to be done the classic way. Ironically, because of its unfulfilled promises, AI unlocked a stage where, quite often, we self-censor due to a lack of means.

Even for a music video, even for just five minutes of film, what truly excites us is the cinematic aspect. We love telling stories in an almost "classic" way—broad as that term may be—and finding a specific narrative grammar to stage a fragment of a film.

Since it is impossible to develop a complete plot with complex narrative arcs in such a short timeframe, the image must contribute more heavily to the storytelling: through its symbols, designs, actions, and pacing. Telling a story purely through visuals is something we find immensely satisfying. It isn't that we hate dialogue, but—to position ourselves as creators "against" the grain—we find that endless talking occupies 98% of current films and series. We’ve reached a point where we only film people talking against blurred backgrounds.

To us, this is forgetting the very essence of cinema: the image. Furthermore, drowning films in an uninterrupted stream of speech ultimately kills the power of dialogue itself. That is neither the approach nor the form that interests us—though that doesn’t stop us from appreciating the work of Tarantino or Scorsese!

Our initial strategy for using AI was to model everything in 3D to guide our original designs and camera angles. At no point did we look to this tool to provide us with designs, concepts, or original ideas. Beyond the fact that AI isn't very good at proposing originality, this is a stage we hold dear; it is an integral part of creative development, sitting at the crossroads of narrative and staging. Everything is interconnected there, and it is impossible for us to delegate that to a machine.

Following our first experience with the Gunship music video, we wanted to proceed in the same way, but by processing video instead of still images. A rough, sketch-like 3D model should have been enough to speed up this normally time-consuming stage, allowing us to focus solely on directing.

However, the limitations quickly became apparent. First, AI is very good at reproducing what it already knows, but handles the unknown very poorly. For a science-fiction film with original designs, this is of no help at all. Of course, there is the option of creating dedicated AI models based on our universe's design (known as LoRAs), but you eventually reach a stage where the necessary upfront preparation is so heavy that using AI loses all its appeal. Furthermore, the more precise directions and elements you give it, the more the AI struggles to produce quality.

For Speed Or Perish, this was our third collaboration with Carpenter Brut. We knew he would trust us. From the writing stage, we develop ideas that will keep us excited over several years.

The best results happen when you give it total freedom: a few words in a prompt are enough to generate something incredible. To control it, you have to provide detailed bases, but this creates a paradox: we pushed the detail of our 3D work so far before "crunching" it through the AI that the result ended up destroying the native quality of our own 3D. We decided to cut our losses and abandoned the AI processing altogether.

Creative opportunity and responsibility In cinema, truly creating your own opportunities to be creative is rare. So far, we have been fortunate enough to design our projects from A to Z, which is an immense artistic privilege. Writing, imagining, and drawing a universe, then transitioning into reality through filming, and finally reaching post-production with such freedom, despite the inevitable constraints, is an exceptional experience. It is an opportunity we create for ourselves by giving our heart and soul. By "exceptional," we mean not confining ourselves to the role of director alone: we build our own sets and create our own digital effects. The absence of an intermediary between the one who builds and the one who conceived the project, between the artisan and the artist, allows for a fluid and organic production. The project can evolve until the very last moment, unlike a traditional workflow where technicians are often disconnected from the artistic decision-making process, and where intermediaries tell other intermediaries what to do. This isn't necessarily a deliberate production choice, like all directors, we would love to be able to delegate much more, but we have turned a constraint on its head to make it work for us and, above all, to make the project feasible.

For Speed Or Perish, this was our third collaboration with Carpenter Brut. We knew he would trust us. From the writing stage, we develop ideas that will keep us excited over several years. Since we know it will require colossal energy in the long run, we must be able to stand by these ideas until the end. We want to be generous, both for ourselves and for the future audience. In Speed Or Perish, the diversity of actions, locations, and pacing ensures that we never get bored, even during production. Spending a year and a half on an eight-minute film might seem crazy, but since no two shots are alike, offering something original makes it worth the effort. We love our craft enough to go the distance.

Finally, our responsibility as Seth Ickerman is to deliver the film requested by our client-musician, but also to honour all the teams who supported us, from shooting to post-production. Beyond the creative opportunity, there is a professional obligation: this isn't a solo project for our own pleasure, but a commission we must fulfill. Motivation or not, we have a duty to move forward and deliver.

Above: The trailer for Blood Machines, the second film from the trilogy.

What were some of the difficulties you faced in pulling this project together?

As we mentioned earlier, making a film means drowning under a tidal wave of constraints, each more imposing than the last. This is the normal process for any production. For Speed Or Perish, we can identify two major difficulties.

The first concerns the shoot. The main challenge was the extremely tight deadline, coupled with the summer holiday period, which meant almost no one was available. Before shooting, there is the animatic phase: a basic but essential pre-visualisation of the film, especially for an action-heavy sci-fi work where everything is virtual except for the actors. We had to know every action, camera dynamic, and lighting direction down to the last detail.

In just a month and a half, we had to write, design, and finalise the initial vehicle concepts to integrate them with the constraints of the set.

In just a month and a half, we had to write, design, and finalise the initial vehicle concepts to integrate them with the constraints of the set. For instance, we had to define exactly how the pilots would sit or the dimensions of the cockpits, which are concrete elements for the actors to work with. In parallel, this month and a half was used to organise logistics, work on costumes, and assemble the crew and cast. The launch of this film was a lightning-fast acceleration, an absolute state of emergency where everything had to click into place to make the shoot possible. Fortunately, this phase is very stimulating, and being surrounded by long-time collaborators allowed us to put all the odds in our favour

The second difficulty was the realisation that AI would not be our ally. Bit by bit, we realised that AI is very difficult to control, erratic, inconsistent, and, above all, mediocre in quality (given the state of the technology in 2025). We kept it in our toolbox primarily for technical passes and a few scattered elements, but overall, if not entirely, the film was produced using traditional methods.

We thought we could gain speed and ambition through AI; from that perspective, it was a failure. However, we refused to sacrifice our visual or directorial ambitions. We chose instead to take the necessary time. Carpenter Brut may have gritted his teeth, but he stood by us and accepted the delay.

How long was the shoot and what was the most challenging aspect of the project?

Considering that the film took a year and a half to make, the actual duration of the shoot will seem almost negligible!

We shot for four days at Studio Lamy, an hour outside of Paris. For us, it’s the ideal location: having filmed Blood Machines there, we were able to take all the necessary time beforehand for prep, such as building the pilot seat and the engine water tank in the days leading up to the shoot. Since we all knew each other well, we managed to sustain very intense workdays. We succeeded in "bagging" everything we had planned, which is quite rare! A huge thanks again to the crew for keeping up with the pace, that’s one of the advantages of a short shoot.

A shoot is always a race against the clock, but this wasn't our first rodeo. We now have a certain expertise in planning and, above all, in preparation, which is the key to everything. The shoot itself wasn't particularly difficult; everything went like clockwork, in an atmosphere of efficient and pragmatic excitement.

The real difficulty was post-production, without the support of AI. We had to figure out how to pull off such an ambitious project using traditional methods.

The real difficulty was post-production, without the support of AI. We had to figure out how to pull off such an ambitious project using traditional methods. The shot count was staggering, and each one was incredibly complex, with multiple layers to mix: isolating the actors from the background, integrating them into their 3D seats, blending the various environments (the city, the track, the numerous vehicles), and adding smoke or explosion effects. But all of that is just the final stage, what we call "compositing" which required three to four months of work. The bulk of the heavy lifting, which took a full year, was the 3D modelling and texturing of the environments and vehicles, followed by their animation and camera movements.

Abandoned by an AI that was still too primitive, we had to push the quality standards of our 3D work even further. Fortunately, we were able to count on the support of "Trimaran VFX" (who have been with us since Blood Machines); we performed all the rendering on their render farm. No fewer than 780 3D projects were processed over 20,000 hours, totalling 80,000 frames. This was an invaluable contribution at the exact moment the AI threw in the towel!

In any case, the pace was gruelling. A year and a half of working at a sprint, without stopping until the very last minute before delivery. Once again, post-production truly lived up to the name: Speed or Perish!

Above: Seth Ickerman's music video for Blood For The Blood God from GUNSHIP featuring HEALTH.

What have you learned during the process of making the film?

The answer will once again focus primarily on AI. How do we tame this monster that is currently devouring everything? AI arrived boasting that it could accomplish anything with a single click, and that is still how most people perceive it. That’s precisely why everyone hates it. Yet, that isn’t the case at all: using AI requires time and an immense amount of work, without ever sidelining human relevance. It forces you to rethink your methods and processes, and, above all, to lower your expectations regarding its so-called "superpowers."

To achieve consistency, control, and cinematic continuity, using generative AI to produce a film’s imagery entirely is not yet feasible. For now, it’s a gadget that delivers incredible results as long as you don’t expect anything specific. But as soon as your ideas are set, the tool becomes unusable. AI has entered the filmmaking toolbox, but it remains just another tool. Beyond the issue of control, professional standards demand a level of quality and resolution that generative AI is not yet capable of providing.

To achieve consistency, control, and cinematic continuity, using generative AI to produce a film’s imagery entirely is not yet feasible.

Algorithmic technology is evolving rapidly, but at the same time, the long-awaited tools designed to eliminate truly thankless tasks are nowhere to be found. It’s only a matter of time; we all know that. For now, we are all searching for ways to wrap our heads around this "baby monster." AI sometimes allows us to save a little time or go a bit further more quickly, but it remains confined to the details.

Ultimately, to our great astonishment, we managed to finish the film using traditional methods after all. Granted, we had to cut the entire opening of the film introducing the characters, but this mission—which seemed impossible to achieve through classic means, turned out to be doable. We are the first to be surprised! In a sense, having beaten the machine this time around is an interesting feat, especially in these times. You have to savor the victory, because it likely won't last.