Gilles Retsin is a London-based architect and designer investigating new architectural models which engage with the potential of increased computational power and fabrication to generate buildings and objects with a previously unseen structure, detail and materiality. His work is interested in the impact of computation on the core principles of architecture – the bones rather than the skin. The practice has developed numerous provocative proposals for international competitions, and is currently working on a range of schemes, among them a 10000 m2 museum in China. His work has been acquired by the Centre Pompidou in Paris, and he has exhibited internationally in museums such as the Museum of Art and Design in New York, the Vitra Design Museum in Weil-am-Rhein, Design Exchange Toronto and the Zaha Hadid Gallery in London.  He has been invited professor at the Texas A&M University, and has lectured and acted as a guest critic in numerous universities internationally. His practice, Gilles Retsin Architecture, will be constructing a pavilion at the upcoming Tallinn Architecture Biennale in Estonia.    

Gilles Retsin studied in Belgium, Chile and the UK, where he obtained a master’s degree from the Architectural Association’s  Design Research Lab in London. Gilles gained professional experience while working in Germany and Switzerland, an architect with LAVA in Stuttgart, as a project architect with Christian Kerez in Zurich and research-based practice Kokkugia in London.

Recently he was appointed the Program Director of the B.Pro Architectural Design (AD) Master’s program at the Bartlett School of Architecture in London. Within the AD program, together with Manuel Jimenez Garcia and Vicente Soler, he leads Research Cluster 4 (RC4), which focuses on developing design methods for robotic fabrication. A new generation of research initiated by RC4 moves from 3D-printing to Discrete Robotic Assembly, utilizing principally simple building blocks to assemble incredibly complex forms while exploring the possibilities of utilizing robotic fabrication on an architectural scale.


Suncheon Art Platform (2016): 278 large-scale, engineered timber elements combine together into a highly differentiated spatial assembly.


How do you respond when someone asks ‘what do you do for a living’?

I would argue that I don’t do something just for a living – I don’t consider my practice and my teaching as a “job”! I am very lucky that my interest and passion for architecture is actually my “work”. This is somehow a very timely question as well, given the current discussion around Universal Basic Income and the future of work. There is more and more pressure on the notion of work, both from the left, and the right side of the spectrum. Silicon Valley, for example, attempts to really blur the distinction between free time and work, as a strategy to achieve higher productivity and flexibility. Work has to become universally “fun”, a passion that people pursue, almost voluntarily. For them, the model of the architect is probably perfect: someone who works long hours, just because he or she likes it! 

Object Oriented Design, (Strange) Mereology, and Object Oriented Eclecticism are some of the terms one will hear when you speak about your work. Could you expand on how recent shifts in philosophy has impacted your world-view and how this has informed your work, if at all?

My work now has absolutely nothing to do with OOO (Object-Oriented Ontology) – but there is some kind of history. When studying at the AA-DRL, around 2010, I came in touch with Speculative Realism and Object Oriented Philosophy. It seemed initially very provocative to me, and in a way enabled me to start questioning some of the things we were taught before. Essentially – we were educated reading texts about networks, flows, continuity, “nature” etc, typical for the digital discourse, or let’s say the proto-parametric discourse. OOO all of a sudden seemed to offer some creative way out of that – slightly tired – discourse. I thought it had some link to object oriented programming as well, so I understood it as a kind of invitation to start to think about discreteness, objects and parts in relation to computation. Not like physical objects – but operations, procedures. I started following Levi Bryant’s blog Larval Subjects and saw immediately lots of ways in how we could start to re-think some of the assumptions of the digital discourse in architecture. Object-Oriented Eclecticism then came up as a first suggestion how this could become architectural. The idea was to look at another way of establishing radically heterogeneous architectural forms, you would basically define lots of different architectural operations or objects, and then let them interact with each other. Every so called “class” or computational object was an architectural operation. So essentially it would allow you to do a collage, without collaging – you would compose lots of radically different elements together, but they would interact and compute something between them. I ran a workshop in Ghent on this topic in 2012, with Isaie Bloch and Corneel Cannaerts. In this kind of spirit, Isaie and I did a competition for a cultural center in Karosta, Latvia. This competition was meant to really do everything in the opposite way as we were both thought to do. For example, I was forcing an agent-based algorithm to distribute straight lines, and really exercise a lot of top-down control over the pattern. Then the overall design was highly material, and seemed to completely detach itself from the ground – it was not attempting to blend in or establish some kind of continuity. It was a big, discrete, autonomous object. It was somehow successful, we got the second prize and the design really seemed to have resonated with the community. According to Neil Leach, I actually became “famous” for doing straight lines with processing.

At the same time, it quickly became clear that other people had a completely different understanding of OOO, and it became a big thing in the US, supported by a generation of people like David Ruy, Mark Foster Gage, Michael Young, etc. I wasn’t really interested in their take on the subject, it was not computational enough for me – and rather conservative. By that time it also seemed like most of the OOO philosophers themselves had already long moved on, and had become more politically engaged. Through the work of Levi Bryant, who I then also met in person when I was teaching a studio at Texas A&M, I became more interested in the notion of part-to-whole relations and the discrete. The idea of the part seemed to combine very well with an understanding of the digital, and with a political possibility to engage with production and automation. So in a way, the OOO thing briefly kick-started the work, and was inspirational, but has absolutely nothing to do with my work and thinking now.

ProtoHouse. Interior and Exterior Views. Prototype for a 3D printed house, 2011-2012, AA-DRL thesis project. In collaboration with Sophia Tang, Nicholette Chan, Aaron Silver. Critics : Robert Stuart-Smith, Knut Brunier, Tyson Hosmer.


 An early project of yours, the widely-published ‘Protohouse’, seems to have laid a strong foundation for your work to this day. Could you expand on the process, both conceptual and technical, that lead to such a radical design?

The Protohouse is indeed the origin for a lot of ideas that keep resonating in my current work and thinking. It’s important to mention here that the Protohouse was a project developed at the AA-DRL, as a collective effort between Sophia Tang, Nicholette Chan, Aaron Silver and myself. We had Robert Stuart-Smith as studio master, but at that time the whole DRL was an interesting place, with also people like Alisa Andrasek, Jose Sanchez and Marta Male-Alemany around. The Protohouse was a prototype for a 3D printed house, and I still like to say that it was the first one. Before that, no one had really played out the possibility of designing architecture at that kind of resolution that 3D printing allows. The Centre Pompidou in Paris has acquired the model now, really because it’s such an important project. Essentially we were exploring the idea of what would happen if we would have a huge 3D printer -not a robot-, with an extreme resolution. We started working with Topology Optimisation, and were especially interested in the kind of gradient structures it could generate. We then made a kind of heuristic version of the algorithm and started adding a series of operations on top of the data from the optimisation, that then eventually results in this fibrous, high-res mass of interconnected lines. Back in 2011, it was almost impossible to visualise and render the whole thing on our laptops.

I like to point out a few things that are hugely important in the Protohouse, and that keep influencing my work until much later. First of all, the main innovation of the protohouse is the volumetric organisation of material. The project does not have any relation to surface, or continuous topology. It’s really a cloud, a container, full of information, that then gets bundled into series of lines. It has nothing to do with classical notions of geometry, there is no mesh, no continuity in a geometrical sense. This was completely new back then, and since then that is a quality I have been pushing in my work and research. No geometry, no surface, no mesh, just points or lines and a series of relations between them. A dissolved mass. The other important aspect is probably the idea of resolution – the Protohouse all of a sudden generated this kind of image of something that is extremely detailed. The aesthetic experience of this extreme detail without a surface is important. Previously in architecture, the surface would be detailed, folded, curved etc – but this project was detailed in the mass – and really extremely detailed. It’s also funny how historically, the Protohouse arrived just in the middle of the whole 3D printing hype – and it really went all around the world. The whole narrative of the 3D printed house was kind of grafted on this project, and it was used by the media to start a kind of fictitious “race for the first 3D printed house”.

Budapest New National Gallery. Honorable mention for the New National Gallery & Museum Ludwig competition in Budapest, Hungary. In collaboration with Lei Zheng and Lieven Block.


 Your work has since shifted from an interest in continuous fabrication methods to discrete parts. Recent projects, your proposal for the Guggenheim museum in Helsinki and the Suncheon Art Platform for instance, seem to contrast your earlier work wherein the computational and design processes consider one particular continuous material as it defines a holistic overall volume, whereas in the latter multiple materials and structural conditions are articulated and explored.

Are the aforementioned projects really as different as they may seem?

That’s a good question. I actually argue that indeed all the projects are fundamentally the same: they are just non-geometrical clouds of lines – “a bunch of lines”. So basically both the Protohouse, Guggenheim Helsinki and Suncheon are all the same project, just with a different degree of tolerance or scale of the part. They share the same syntax. So although Suncheon is very low-res, to the point that some people even say that it’s minimalist, it still shares the same syntax as the other projects. At the same time this argument is of course completely false, because you could as well say that they are fundamentally different, both in aesthetics, mode of production and part to whole relations! The Protohouse has no notion of parts, the Guggenheim still needs lots of customisation, and Suncheon is really the opposite of these two. Anyway, what is clear is that they are all part of the same agenda, and are fundamentally investigating the same issues and questions.

Guggenheim Helsinki. Proposal for the new Guggenheim museum in Helsinki. 2016. Gilles Retsin Architecture. Rendering: FlyingArchitecture

What prompted this shift in your approach to design and construction?

I mentioned before that I started to become more and more interested in the notion of the element or the part. The Protohouse project essentially had no parts – just operations perhaps. When I started doing projects with the practice, I was forced to think about something that would be buildable, today, without a 3D printer or robot – that would still maintain the same radical algorithmic logic and architectural experience. The Guggenheim Museum in Helsinki came out of this logic – it’s a large volumetric mass of timber elements, which together form a fibrous roof structure. It’s all about these timber pieces basically establishing something with quite a bit of repetition and serialisation. All the parts of the building are defined as separate entities, which do interact with each other – perhaps something left over from the Object Oriented Eclecticism idea. So basically the floor slab is made of a much simpler, rectilinear organisation of elements, and does not have any continuity with the roof or columns. The Guggenheim was in a way a kind of more controlled version of the Budapest New National Gallery project, a competition we did before, and actually got selected before it was cancelled. The Budapest project is again a similar, highly material organisation of thick, volumetric, high-res slabs. Budapest really set out the agenda, maybe more than Helsinki. By basing the building on three large slabs, a bit like Mies van der Rohe’s New National Gallery – the whole thing rejected the idea of geometry and formal articulation of spaces as an agenda. By assuming the most “boring” form: 3 perfectly straight slabs, it’s really saying that from now on the only thing that matters are the parts and their organisation – not the form or space itself. That project of formal and spatial research has been done. This is still the position that causes most of the controversy and debate around my work – I get accused very often of doing modernist slabs. The later projects, like Suncheon, are also continuing that question about the status of the part, but in a more radical way. The part really becomes a kind of digital building block, an element like a Lego, with a specific set of possible connections. This way of working came out of a questioning if it would be possible to make buildings that are physically digital – a kind of criticism towards the project of the digital. But it’s at the same time also really concerned with certain aesthetic questions, compositional issues, etc.


Suncheon Art Platform and 278 Parts for Suncheon Art Platform. 2016. Gilles Retsin Architecture.

The ‘Suncheon Art Platform’ in particular seems to re-examine the archetypal part-to-whole relationship of the architectural edifice, whereas the building is not simply an overall shape constructed of parts, but rather the parts themselves are seen as discrete units which accumulate to form the building.

Yes exactly. So Suncheon is the latest large-scale proposal of the office, and is very much related to a housing project we worked on before in Belgium, called Diamonds House. The fundamental questions behind these projects are really based on what it means for architecture to be digital, or to relate to the digital. I think it’s intellectually and culturally a very important question to ask. Our whole world is rapidly becoming more and more digital. In a way it’s so obvious we somehow, as architects, need to intellectually engage with these changes. We need to absorb them, reflect on them, to remain culturally relevant. I am not at all saying that this is the only issue, but it’s one of them. It’s funny how after 2008 the digital experimentation in architecture has been kind of cast as “socially irresponsible”, and we see a whole backlash against it in many architecture schools, Venice Biennale’s, etc. I would argue that we need to be pro-active, rather than reactive, and take ownership of the tools and ideas of the digital. We need to be propositional, and we’re in a good place as architects to talk about production, manufacturing, the city, climate change – these kinds of things are our domain really. At the same time I do share a lot of the 2008 criticism – the past decades with experimentation in the digital were a kind of formalist mono-culture, which definitely also didn’t really restrain itself with too much social consciousness. I think Suncheon is one of the clearest proposals for how we can re-think and re-design an approach to the digital that is fundamentally different from the previous two decades. It consists of a family of parts, with a few different scales. They’re hollow engineered timber boxes, which can be combined in many different ways. They’re independent of a specific building, but at the same time they are put together into a highly specific building – which divides the site into a series of courtyards and squares. The building is physically digital – made out of large scale discrete blocks, without any notion of surface or topology. There is no mass-customisation – just algorithmic organisation of these bit-like elements. Its material organisation is radically different from any other building. Aesthetically – it does not attempt to create a kind of streamlined, gestural image like many of the digital projects do – trying very hard to convince you that the future has arrived.

Diamonds. Undisclosed project for a multi-family house. 2016. Gilles Retsin Architecture.


While operating as both architect and researcher, how do you mediate between pure computation and design? Between matter and intent?

I would argue that today this boundary between the architect and researcher is blurring, I see them at least as very interconnected. I consider the work I do with the practice as a form of research – and I effectively also publish papers about it. The teaching and research at UCL is then again also really connected to the work in the practice. I think we’ll see more and more architects following this model. Of course there is a practical side to this which makes the mode of operation quite different – the practice needs a certain speed and intensity to develop, whereas the research is much more collaborative and slower paced. The practice is perhaps also wilder, more free, more diverse than the research. When doing research there is a certain kind of long term consistency – a kind of responsibility. The practice on the other hand can develop something more personal, sometimes more funny, sometimes more cynical. You can do an unexpected move, and go much faster. I enjoy that. And yes, you are probably right that it’s more rigorous – you are forced to make quick decisions, to evaluate things immediately.

The Bartlett School of Architecture’s MArch Architectural Design (AD) and the MSc Architectural Computing (AC) programs will collaborate within the B-Pro umbrella. For more information visit


Following up from Alisa Andrasek, you recently became the Program Director of the B.Pro Architectural Design (AD) Master’s program at the Bartlett School of Architecture in London, and you also teach in the same program. The set-up is that of an architectural research lab in which students work in teams with specific agendas.    What might a prospective student expect to gain from the program?

B.Pro or “Bartlett Prospective” is an umbrella of intensive, forward-looking post-professional 1-year masters programs at the Bartlett, founded by Frederic Migayrou and Andrew Porter. Currently, we offer MArch programs in Architectural Design (AD), Architectural Computing (AC), Urban Design (UD) and an MRes in Digital Theory – with Mario Carpo. AD is a large program, with roughly 130 students and 8 so-called Research Clusters, all with different agendas , but bound together by an interest in design and technology. The program has a very young and experimental faculty – working on different interests ranging from robotics to materials and craft, to bio-tech and artificial intelligence – but all within a design context. In that sense, the program really equips students with some of the most important skills for the coming decades: design, making, and computation. Our students enter the program often with a very classical education in architecture, but graduate with a completely different mindset, and the ability to work and participate in different disciplines. They learn advanced design skills, programming, modelling, robotics but also traditional craft – and at the same time they write theoretical papers and develop a conceptual background for their work. The year finishes with a big exhibition in September. This coming year we’re also experimenting with a new setup where our AC students will collaborate with the AD and UD ones and vice versa. This is probably one of the most exciting developments in the school now. The idea is to really build AD/AC into a platform for design, computation and critical thought – a place that would attract prospective architects, product designers, but also engineers and computer scientists.


Could you expand on the main interests and agendas of your Research Cluster?

So I lead Research Cluster 4 ( RC4) , together with Manuel Jimenez-Garcia and Vicente Soler. The past few years we have been developing what we call “Discrete Design Methods” – design strategies that step away from the decades old idea of continuity and mas-customisation, and instead propose an agenda based on the notion of the part. We argue that continuity is not even digital – and is just a mere automation or “computerisation” of otherwise manual actions. We’re trying to establish a whole new framework and syntax for architecture, computational design and digital fabrication – that is not just “computerised”, but fundamentally digital. For the moment we have two main research strands; one is discrete methods for spatial printing, and the other one is based on discrete additive assembly of building blocks. Previous years we tested everything on the scale of furniture, but this year we have started to move into larger scale assemblies, and 1:1 prototypes.


What do you have in store for the coming academic year?

Next year we will continue our research into large-scale discrete fabrication – working with 1:1 prototypes and developing more in-depth design strategies for the architectural scale. We’re moving away from the furniture pieces that we did initially – we’re ready to go into a more architectural discussion! Our new facilities at Here East also give us the chance now to really experiment with larger scale work, and start to bridge towards the industry. 


SpatialCurves and Robotic plastic extrusion of continuous systems. Robotically fabricated Chair Protoype v3.0. The Bartlett AD-RC4, 2014-15. Team: Curvoxels. Tutors: Gilles Retsin and Manuel Jiménez García. Photograph.

INT: Robotic Building Blocks. Robotically fabricated chair consisting of complex prefabricated building blocks. The Bartlett AD-RC4, 2016. Students: Zoey Tan, Claudia Tanskanen, Qianyi Li, Xiaolin Yin. Tutors: Gilles Retsin, Manuel Jiménez García with Vicente Soler. Photograph.


Fossilized and Robotically 3D printed Concrete Table Prototype. The Bartlett AD-RC4, 2014-15. Team: Amalgama. Tutors: Gilles Retsin and Manuel Jiménez García. Photograph.


In the context of computational design and fabrication, what do you think the future holds for architecture?

It’s an exciting time for fabrication – we see how research institutes like the ETH NCCR attract significant government funding for Digital Fabrication. Also the Bartlett, for example, is expanding its facilities with Here East, a massive interdisciplinary making space near Hackney Wick. On the other hand there are also architects themselves who are starting their own factories or workshops, like Jelle Feringa with Odico, Mustafa El Sayed with Automata Technologies, Daghan Cam with Ai Build or my colleague Manuel Jimenez with Nagami. At the same time there is still a lot of work to do on how we deal with digital fabrication – for example the idea of the Fab Lab, which was very influential in architecture, urgently needs an update. Fab Labs were a first idea to start distributing new modes of production, combining it with some form of education – but as people like Evgeny Morozov have explained, it suffers from a kind of arts and crafts mentality – and in many ways is a typical neo-liberal project. I think we should, as architects, actively work on how a future driven by digital production would work. This is a very big project – which takes computational design and fabrication outside of its politically ignorant position. As Paul Mason and others have argued, digital production forms the basis for a new post-capitalist society. So it’s important we don’t only think about how to make some nicely crafted chairs, but also consider all of this as a wider cultural and political project – which allows us as architects to contribute to larger societal questions.

Tallinn Architecture Biennale. Winning competition entry to be constructed during the Tallinn Architecture Biennale, Estonia, 2017. Gilles Retsin Architecture with Nous Engineering and support from the Bartlett, UCL.


 Regarding your practice, Gilles Retsin Architecture, would you comment on the process that takes place when selecting a project to pursue?

In the first two years of the practice, we were mainly doing lots of competitions, as a means to quickly cover a lot of ground and experiment with different typologies and scales. It was really good to get some of the work grounded after the wild experimentation in academia. The competitions really helped crystallizing and catalyzing a lot of ideas. However, after two years of doing them, we started to try to do more self-initiated work and research, which lead to a lot of interesting work. The office now has a pretty clear agenda, and we’re getting ready to slowly start to scale up a bit and push more seriously for building something. Before, we were pretty relaxed with not materializing things, the main aim was really to explore an agenda first. Of course, this is still in development as well, and should keep developing further over time.

A lot of the work we do is also self-initiated – there is a whole pile of research and design work waiting to be tested. This is very important – to keep producing work and ideas, even without a client or anyone asking for it.


As a young designer, who or what had an influence on your work? Who’s work is currently on your radar?

Most definitely my experience working in Switzerland with Christian Kerez, and then the subsequent move to London to study at the DRL. The combination of these two experiences was really important, it gave me a kind of overview of different approaches and discussions. I think it’s a very important to look outside of your own circle or let’s say “community” and see what other people are doing. It’s really interesting to see the discrepancy between for example different academic discussion and then the discussions that happens among certain circles of practicing architects. Engaging and crossing over between these multiple conversations is very important for me. When I started the office it was really the idea to do work which would not be marked as “digital” or only speak to that group of people – I wanted to have something that floats above these discourses. And in a way that has worked out, projects like Suncheon and the Diamonds house are successful in that sense because they manage to be both part of a critical digital discourse, a very propositional approach – and at the same time also appeal to a large community which is maybe not at all interested in the theoretical ideas behind it. So in that sense the work that I look at and follow now is also quite eclectic – I like to use Instagram and Facebook to get a sense of what is going on in different circles and learn about their work. I may disagree with what they do, but still appreciate it and learn something!


Karlsplatz II . Proposal for a museum at the Karlsplatz in Vienna, Experimental Architecture Biennale, Prague ( 2015).



What advice would you give to young up-and-coming architects and designers today?

I am probably not the right person to give any kind of career advice or so – as I am myself part of this category 😉 I do want to maybe throw some ideas out there, about what I think me and my peers should be doing. Essentially I think we need to become much more propositional, and start to question much more as a generation what we want to do with technology and how we respond to the current challenges. I think we have to be really critical of the current generation of techno-phobic people, who are running a lot of the schools and big architectural platforms now. People like Alejandro Aravena, who are claiming the agenda of social change- but at the same time are funded by oil companies, and advocate a conservative design agenda. I think we have to break this narrative, and make it clear that technology and social/political action are not opposed to each other: contrary to that – it’s exactly the combination of these two that is the solution. This is what the whole “post-capitalist” movement is about. As a generation we need to claim that territory, and start to engage with it through work – and I mean both design and theory. Let’s start to think about digital tools and fabrication, not just as a means for experimenting with funky shapes, but also as means for change. Again – it’s important to emphasize that I am not at all rejecting an adventurous and creative design agenda here, it can go hand in hand. The worst thing would be an explosion of “wiki-houses” – which is a project I really admire, but in all fairness is not very interesting in terms of architecture and design. I hope that some of my projects demonstrate the idea that you can both look into new architectural ideas and engage with social agenda’s through the idea of production. So, in brief, I would hope my generation and peers would be able to take a mature position in relation to technology, understanding it as a tool for creativity and change, and start a form of collective practice and research that helps us prototype and think about a post-capitalist world. I hope this form of practice would be really creative and diverse -with multiple different approaches and opinions to the technology.


What’s next for Gilles Retsin?

There is a lot going on, we just won the competition for the Tallinn Biennale Pavillion, which will be built by September. That’s a big step for the work – as this will be the first discrete building to be realized. There’s also a series of other projects in the pipeline, ranging in scale from a museum in China to prototypical housing blocks. I am also working on a series of drawings and paintings, and writing a lot. Together with Manuel Jimenez, I am working on a series of commissions with the 3D printed and robotically manufactured furniture. We’re also developing a software for discrete design and fabrication, and there’s lots of conferences, books, and articles scheduled.


Blokhut (2015), Study for a private house in Wetteren, Belgium, 2015.

Leave a Reply

Send this to a friend

Subscribe to our weekly newsletter.. It's Free !
Email confirmation is required. Please check your Junk E-mail folder just in case !
Subscribe to our weekly newsletter.. It's Free !
Email confirmation is required. Please check your Junk E-mail folder just in case !