Head of Research at Composite Films

Hi Camille, and welcome to this conversation, it’s truly a pleasure to have you here.
Looking at your academic background, there’s a clear and consistent interest in history, collective memory, and the audiovisual world.
How much has your education, between the Sorbonne and INA, shaped the way you approach your work today?

During my Master’s in American History, I wrote a thesis on the street children of New York in the 1920s. For that project, I carried out extensive iconographic research and worked with a wide range of materials, including photographs by Lewis Hine and early Edison films. I was struck by the narrative potential of these archives — their ability to document a
historical moment, of course, but also to evoke a whole world: the texture of a time, the mood of a city, the invisible stories behind the frame. Thanks to my background in history, I learned to question images rather than take them at
face value — to dig beneath the surface, to read between the lines, to consider what’s shown and what’s left out. At INA, I gained a more technical understanding of how audiovisual materials are made, structured, and preserved. That dual training — historical and technical — gave me the tools to work with archival images in a way that’s both critical and creative. I believe this combination is essential when you’re trying to tell meaningful stories with footage from the past.

Today you’re Head of Research at Composite Films.
Can you tell us what your role involves exactly? What does a visual researcher do in the context of documentary filmmaking?

Composite is both a colorization studio and a documentary production company. About three years ago, they acquired Archive Valley — a platform designed to help filmmakers and content creators find the archival footage they need. At the time, I was Head of Research at Archive Valley… so I came as part of the package!
At Composite, my role is now divided into four main areas:
Archival consulting:
I continue the core mission of Archive Valley by helping filmmakers, researchers, and producers source the footage they need — whether it’s for historical accuracy, emotional resonance, or narrative structure. I assist them with everything from locating rare materials to clearing rights and understanding licensing terms. I also maintain and expand our
global network of archive providers and researchers, which is essential in a field so reliant on collaboration and trust
Colorization research:
For our historical colorization projects, I conduct in-depth research to ensure every detail is accurate — from the exact shade of a military uniform to the color of a street sign in 1940s Paris. Colorizing history requires a lot of rigor, and it’s my job to make sure we respect the truth of the images while giving them new life.
R&D and AI tools:
I’m also involved in research and development for an AI-powered tool we’re building to streamline the colorization process. The idea is to speed things up while maintaining — or even improving — historical accuracy, by training the tool with well-documented visual references and expert input
Documentary development:
Finally, I work on developing new documentary projects for Composite. I handle all the visual and archival research, from early concept to final cut, making sure the narrative is grounded in compelling and meaningful footage.
It’s a very hybrid role — part historian, part detective, part storyteller — and that’s exactly what I love about it.

A visual researcher plays a crucial role in shaping the visual identity and narrative depth of a documentary. Our job is not just to find images, but to find the right images — those that will carry emotion, context, and meaning.
We start by understanding the director’s vision and the story being told. From there, we dive into archives, photo collections, newsreels, institutional footage, sometimes even amateur films, to uncover materials that resonate with the themes and characters. It’s a mix of investigation, intuition, and historical knowledge.
But research doesn’t stop at finding footage — we also verify sources, trace rights holders, negotiate licenses, and make sure everything can legally and ethically be used.
In many ways, visual researchers are like invisible co-authors: we help documentaries find their images — and sometimes, the images reshape the story.

You’ve moved from production roles to those more focused on research and building narratives through archival images.
Was there a specific project, or even just a moment, when you realised this was your path?

I actually started my professional life working with archives — they’ve always been my first passion. Then I moved into production thanks to a great opportunity, and over time, I gradually found my way back to archival work. In a way, it felt like returning home.
That experience in film production has been incredibly valuable for the work I do now. It gave me a much clearer understanding of how a film is actually made — the workflow, the constraints, the constant balancing act between creative ambition and practical limitations. It also taught me how to better communicate with directors, producers,
editors — to speak their language, anticipate their needs, and offer solutions rather than just images.
So while I may have come back to archives, I bring with me everything I learned along the way — and I think that makes me a stronger, more empathetic researcher.

What’s the part of your work that you enjoy the most?
The one that still gives you satisfaction, even when deadlines are tight and the process is demanding?

The part of my job that I love the most is immersing myself deeply in a subject and progressing through the research process. I really enjoy playing detective — hunting for unseen or rare materials and discovering new angles on a topic.
Even under tight deadlines or pressure, those moments of discovery give me real satisfaction. It’s like finding a hidden treasure that can shift the whole narrative or add unexpected depth. Sometimes it’s a tiny detail — the color of a sign, a background gesture, a forgotten face — but it can change everything.
I also believe it’s incredibly important to connect with specialists and fellow researchers. Exchanging ideas with them opens new windows of understanding, suggests valuable readings, and points me towards archives or resources I might not have found on my own.
This kind of collaboration and intellectual emulation is vital, especially since research can often feel like a solitary pursuit.

Over time, you’ve worked on very different types of projects, across eras, countries, institutional and private archives.
Do you still find yourself surprised by a reel, a photograph, a film?

Absolutely — no matter how much experience I gain, I’m still regularly surprised and moved by what I discover in archives. There’s always something new or unexpected: a reel that captures an intimate moment, a photograph that reveals a hidden story, or footage that brings a forgotten era vividly to life.
Working across different eras, countries, and types of archives has taught me that history is never fixed or fully known. Archives are full of surprises, and each discovery can challenge or deepen our understanding. That sense of wonder is what keeps me passionate about my work — it reminds me that archives are living treasures, full of stories waiting to be told. I remember coming across a powerful collection of footage documenting the Rwandan genocide — with incredibly rare testimonies that hadn’t been widely circulated. Another time, I spotted a very young Joan Baez in the crowd at the Newport Folk Festival in the early ’60s — just a few frames, but full of emotion and symbolism. Moments like those
remind me how vivid and human archives can be.

I’m fascinated by that moment when an archival image starts to “speak” and finds its place in a story.
Has it ever felt like a sequence chose you, rather than the other way around?

Absolutely — I’ve often had the feeling that a sequence chose me, rather than the other way around. That’s the magic of archival work: sometimes a piece of footage reveals itself and suddenly shifts the direction of a story. It has a voice, an emotional charge, a presence — and it demands to be at the center.
For me, the best way to build an archival documentary is to let the narrative emerge from the footage itself, not the other way around. When you start with the images — really watch them, listen to them — you allow the archive to speak with its full power. If you do the opposite, writing the film first and then looking for images to fit, archives can easily become mere illustrations, secondary to the story rather than essential to it.
When I was working at Archive Valley, one of our missions was exactly that: to uncover powerful archival collections sleeping in vaults and bring them to the attention of filmmakers — giving them the chance to build narratives from the footage itself. I remember incredible collections in regional Cinémathèques in France showing May ’68 from a local, deeply personal perspective, or unique footage documenting the early years of the AIDS crisis. These archives don’t just support stories — they are the stories.

Is there a film, a scene, or even a single frame that you’ve become particularly attached to over the years?

In recent years, we’ve heard a lot about artificial intelligence, including in the archival field.
How do you think this technological shift is changing (or might change) the work of visual researchers and the documentary industry more broadly?
And more specifically: what are your thoughts on AI-generated images and videos that try to mimic archival footage?

AI is already having a significant impact on the archival and documentary fields — and it’s just the beginning.
From a research perspective, AI tools can be incredibly helpful in processing large volumes of data quickly. They can assist with image recognition, metadata extraction, or even automatic transcription and translation. For visual researchers, that means we can spend less time on repetitive tasks and more time on the deeper, more creative part of the job: interpreting materials, finding connections, and curating meaningful narratives.
At Composite, for instance, we’re developing an AI tool to help accelerate the colorization process. But the goal isn’t to replace expertise — it’s to support it. We’re training the tool using historically verified references, so it becomes a partner to the human eye and not a substitute. For me, that’s the key: AI should remain a tool, not a decision-maker.
Now, when it comes to AI-generated images or videos that mimic archival footage, I’m more cautious. These synthetic images might have artistic value or be useful in very specific contexts — for example, to visualize events where no images exist, as long as they’re clearly labeled. But they also raise serious ethical questions about authenticity,
memory, and trust.
Archives carry emotional and historical weight precisely because they are real fragments of the past. If we start blurring that line without transparency, we risk undermining the unique power of genuine archival materials — and with it, our relationship to history itself. So I think AI opens exciting possibilities, but we have a responsibility to use it carefully,
especially when working with collective memory and public trust.

What advice would you give to someone who wants to follow a similar path today?
Where should they begin?

A solid education is essential. Being a good archival researcher isn’t just about being a great detective — it requires a diverse skill set. You need to understand copyright and licensing laws, have solid production knowledge, and know how to navigate the complex rights attached to archival materials. Negotiation skills are also crucial, as is a strong
familiarity with collections and archives from around the world.
Don’t underestimate the value of building a network of fellow researchers. This can be a solitary profession at times, and sharing tips, sources, and experiences can be incredibly enriching and helpful.
Platforms like Archive Valley are a great resource — we provide high-quality training and access to a vibrant, international community of researchers. The APA (Archive Producers Alliance) and FIAF (International Federation of Film Archives) are also excellent starting points. They offer a wealth of academic resources, curated reading lists, and webinars. But above all, you need a passion for storytelling. At the end of the day, we’re not just sourcing footage — we’re helping to bring history to life.

Thanks again, Camille. I look forward to seeing where your research takes you next, and hopefully crossing paths in the archives.