Cerrillos, south of Santa Fe. I moved from NJ for health reasons several years ago and rent a small adobe cottage on a mesa. It's actually the same location mentioned in Nader Khalili's book Sidewalks On The Moon, where he briefly stayed on his way to Los Alamos Labs to present the idea of adapting the ancient Persian technique of fired earth block construction to solar-thermal vitirified masonry for lunar settlement construction. (NASA later moved on to his SuperAdobe concept based on coiled earth bag structures)
I had considered the Fab Lab approach because resources are so scarce out here that I felt it necessary to try to seek combined support from the MIT program, local schools, the state, and the native government. And I wanted to sell the native communities on investing in their own potential industrial self-sufficiency as well as education--preparing the young for entrepreneurship within their own community, not mere jobs somewhere else. But some places are fertile ground for this sort of thing and some aren't. Santa Fe may be left-leaning and environmentally conscientious, but it's not exactly what you'd call progressive. I think the general demographic may just be too old. Most Santa Fe locals can't tell you who that Paulo Soleri their city's public theater is named for is.
Inflatable pressure shells/hulls are a great way to go for modest scale habitats because they can be used in many ways and have that key high potential repairability and easy replaceability that are the chief failings of the more traditional outpost architecture. They are also something well suited to on-location industry--albeit at a relatively high level of sophistication--because, in space, the essential limitation on manufacturing is that you can't precision-fabricate things in the ambient environment too easily and you can't make anything inside a habitat that can't be fit through an airlock. (or at least some kind of larger temporary pressure enclosure) That's a fundamental limitation that's very commonly overlooked.
The limitations of inflatable pressure hulls are that they really can't be load-bearing or put under a lot of tension (like a tensile fabric roof system), and they are fabricated to be contiguous or monolithic and so start getting unwieldy at large scales. So you tend to have to combine them with other structural systems. In TMP2: Asgard I discuss many uses of these kinds of hulls in combination with space frame systems and other structures for both habitats and built-on-orbit spacecraft. The simplest form is, of course, the TransHab-type hull which combines many layers of material to create both a pressure shell and a Whipple shield protection system. But because there are so many material layers the TransHab is limited in how compactly it can actually be packed, which tends to limit the maximum size of the unit habitat shell. Generally, this is so far beyond the size of habitats built to date that it might seem moot. (one of the objections voiced to the idea of converting Shuttle fuel tanks to habitats was that NASA engineers simply couldn't imagine any practical use for that much space. The currently planned Bigelow BA2100 inflatable module will have more than twice the working volume of the entire ISS--in one module. In the future, we'll be regarding the ISS as a radio shack) But that type of hull also has limits in radiation shielding. It's limited to the LEO environment because it's difficult to integrate any greater shielding volume into that structure.
Thus I explored a concept called EvoHab; a composite hull system that separates and modularizes the functions of pressure containment, structural support, and shielding and could transition in scale from the small spacecraft to city-sized megastructures. At modest scales, EvoHab hulls use a dedicated contiguous inflatable pressure skin which is deployed within, or around, a space frame structure, attaching to a surrounding frame by welded/glued-in-place node-point snap-in connectors. These attachment connectors might have 'pass-through' connectors allowing internal frame structures or other elements--like lighting--to be added and communicate load/tension to the outer frame but not to the pressure skin itself. Outside the pressure hull could have modular shielding panels of any desired thickness attached, simply, by velcro (assuming a relatively small habitat) or, if using that external space frame, retrofit to that frame. Many functional elements would also retrofit to that frame and could use the interstitial space behind them for shielded utilities routing. One key use may be heliostats that gather light and pipe it through fiber optic couplers to provide natural illumination of the interior, making the hull virtually translucent no matter how heavily shielded and, given the right optics, potentially virtually transparent. For very large structures one would go to a built-up pressure hull where one assembles a surrounding frame, adds shielding to create a sheltered unpressurized working environment inside then you apply inside foundation panels onto which something like a fiber-reinforced epoxy is applied to create a contiguous pressure-tight shell--rather like how one makes conventional housing walls by putting up sheet rock on a wood frame and covering it in plaster. So, altogether, you have a strategy to make perpetually repairable, changeable, and expandable pressure hulls fully integrated with internal and external space frame structures. This makes things like building space habitats literally from scratch and assembling manned spacecraft on-orbit--now considered largely impossible--a relatively straightforward task.
For surface habitat applications you have many ways to use these inflatable shells. The simplest is to simply deploy them in some kind of excavation affording protection from the natural rock. Excavated habitats are the simplest way to employ ISRU to permanent habitat construction while being well suited to robots. Today's roadheader excavators are practically telerobots already, being largely automated, laser guided, and almost as precise as a CNC. Excavated habitats can also, given sufficient density of strata, eliminate the need for the inflatables allowing structures to be pressurized with just the addition of modular bulkheads and perhaps the application of reinforced epoxies and ceramics. TMP2 goes into this in much depth with the addition of retrofit attachment grids based on rock bolts that allow installation of panel and framing systems for interior finishing. The catch is, of course, that nature doesn't always provide the easily accessible strata you need in just the location you might need so, long-term, you have to think about this in combination of built-up construction that produces largely the same sorts of spaces by other means. There are many viable approaches.
One of the simplest is the earth-bermed quonset or arch structure. You assemble an arch enclosure from rolled corrugated alloy--just like the common farming structures--cover it in loose regolith with the help of soil stabilizer mats, and you have a simple artificial cave to deploy your pressure shells inside. The metal rolling hardware for making these arch sections is often portable today so a variation of that could use material supplied from Earth in compact rolls--though this would be rather heavy. Rolled sheet alloy is also a likely form of material from later ISRU.
Another approach to this using less imported material is the SuperAdobe scheme that was invented by Nader Khalili and explored by NASA some years ago. That method uses earth bag tubing with velcro strips that are filled with loose regolith mixed with a stabilizer then coiled up to form arch and dome shapes. The filling and coiling can be combined in a more-or-less continuous process. The end result is, again, a simple cave you can deploy inflatables in. The catch is that this isn't really suited to current robotics. It's a manually complicated process and if you're relying on human labor, you're likely limited to very small structures because of how hazardous that is. A variation on this concept could integrate the 'earth bag' elements into the inflatable shell so that, over-pressurized or using high-pressure ribs, it would function as a pre-form while filling the outer earth bag elements (most likely concentric ribs similar to the form for SuperAdobe) which then become a self-supporting shell when complete. The question is how you transport your material into the earth bag segments and compact it if they are relatively long and attached to the hull.
Assuming you have developed some kind of regolete (regolith-concrete/geopolymer) material or a viable regolith sintering technique, the simplest way to make large rigid shells from it is by the mound forming method most dramatically demonstrated by German engineers in WWII for the construction of buried bomb-resistant aircraft hangers and bunkers. Basically, you use earth moving equipment to pile, compact, and shape a mound of granular material in the desired shape of your structure, cover that in a reinforcement grid, then pour your regolete mix on top of that to form your rigid structure. If you're using a sintering method, you would just spread and sinter successive layers of material to build up the rigid shell. You are limited in the degree of slope that can be tolerated without additional forming structures, but you can use any combination of arched and dome shapes. These can then be earth-bermed for additional protection. Once complete, you then dig out the inside material and, again, you've got a nice cave to deploy your inflatables in or possibly seal with reinforced epoxy and ceramic. This is a technique well suited to robots but the limitation is that it's pretty high-energy. You're moving around a lot of material, even if you don't need especially fancy tools to do it.
Of course, conventional slip-forming, climbing form, and precast concrete techniques would also work just as well if the regolete material can accommodate it and means of automating them devised. Recently there has been a lot of hype on the Internet about the prospect of 3D printing space habitats but that's really just a more elaborate way to produce the same thing as these other methods; a heavy rigid shell of regolith-derived material that you can deploy an inflatable skin inside or maybe use as the foundation for a reinforced epoxy or ceramic layer. Given the likely simplicity of the shapes and thus the simpler methods of automating their fabrication, one wonders if this is overkill. For an application like this, today's common 3D printer designs have a big problem. They all use linear/scalar positioning systems which, when you get to working at the scale of buildings, start getting ridiculously bulky because they have to surround the whole working volume. Large scale 3D printing really wants hardware based on radial/vector positioning systems where a multi-stage arm extends from a small space, resulting in a much more compact and portable system that might even just sit on a truck. We haven't actually seen this in any form of positioning in any 3D printer yet even though its common in robotics generally-possibly because of the greater complexity of the underlying mathematics. But given a relatively simple shape like a dome, far simpler machines are adequate. Imagine a radial boom with an extruder of some type. There used to be an entrepreneur in the US offering super-efficient circular concrete homes made with a simple machine like this. The machine would just spin this extruder boom around drawing out concentric circles or radial lines of concrete and building the walls up layer by layer. They would then top it off with radial beams or vigas and composite roofing and apply whatever finish was desired. Heavily inspired by the Earthship concept, they featured rooftop greenhouses, solar power, and the like.
The EvoHab concept would also work more-or-less directly for surface structures but is not as well suited to early ISRU because it's components need more sophisticated production. The exact same system used in orbit could be used in the surface environment, just with different shapes suited to the gravity and with an option to make some elements from some local materials. One could use a SuperAdobe-like approach to make foundation and panel elements using fabric 'bags' with a 'quilted' structure to provide flat shapes. These might have quick connectors built into their corners to plug directly into the node points of a surrounding space frame or 'hand' mounts as used with planar glass systems. Or, if more sophisticated local fabrication were possible, a form of foamed regolete or sintered corrugated panel could be used to make plug-in shield panels much like those used in orbit. Though not as well suited to early ISRU, these kinds of structures would be well suited to telerobotic pre-settlement because the structural elements would be small and light in size, potentially easy for robots to handle, and well suited to dense packing and 'acceptable yield' logistics where you accept a certain tolerable failure rate in transport (sea launch, rough landers) because the value of payloads is kept modest. In TMP2 these kinds of structures are employed for Telebase service shelters used to shelter robots under repair and assembly and temporary manned 'construction camp' habitats used while larger shelters based on more local materials are produced. (though generally this is a concession to anticipated impatience. As the Telebase takes shape it would tend to compel increased desire for human visitors even if, technically, they probably won't be needed until the larger permanent structures are complete)
There are a lot of interesting approaches that can be explored with this, but not too many people actually doing it because the intermediate-sized habitats and the use of telerobotics are just overlooked in most space settlement proposals.