A potpourri of ideas that could integrate with the concept you outlined above:
Live-in business hatchery. The young professionals mentioned by Alberto, or rather young entrepreneurs, might also suit as a second (minor) target group. Because there will be quite some young entrepreneurs tired of living with their parents, and looking for a nice but very affordable home. If that happens to be a live-in startup incubator, all the better.
Pick to pay. There is a social issue in Kathmandu in that coffee is considered a "high class" beverage, quite unaffordable for ordinary people. So, a nice opportunity to deconstruct that. Idea: have 1-2 special tables in the café where people come to chill / hang out and get coffee and seats for free, as much as they want, if they sort a bowl of coffee beans that comes along with every cup of coffee. The desks would be clean bright desks suited for sorting, with bright lamps, giving a reason why not everyone in the café can take part in this scheme. There is not a hygiene problem at all, since all germs will die when roasting at ≥170 °C.
Coffee picking as a service. Potentially, the "pick to pay" concept even makes business sense, and can be commercialized by offering Kathmandu's coffee traders and exporters to sort their coffee for a fee. Coffee has to be sorted to become high quality or even speciality quality coffee, and there is probably not a single optical sorter machine in Nepal for that. So the only option is hand sorting, as for 80% of the world's coffee. (Until our open source optical sorter is developed and built, that is.)
Align with a makerspace. A good proposal could be to find a space close to an existing makerspace / hackerspace, like Communitere Nepal. So people get the best from both worlds: nice coffee and good Internet in the co-working space, accommodation in the co-living space, but also a workshop with tools etc..
Hosting international volunteers. A good secondary target group for the accommodation, because they mingle well with students, are international volunteers (for example from AIESEC). They have usually less of a problem than students to afford room rent, so can cross-finance others' space rent a bit.
Get the co-working space for free. As we calculated when looking at the FOAM candidate space in Brussels, it is possible to get the co-working area "free", covered by the rent of people in the co-living space. This relieves it of the pressure to make a profit on its own, which can greatly enhance user experience by being much less commercialized. And from that relaxed, enjoyable setting, a lot of popularity and other options to make a profit may come, who knows.
Don't waste the coffee smell 
Coffee smell is precious and almost universally loved. After roasting, the coffee has to cool down, and cure. That can be done in such a way that the smell goes into the guestroom. Maybe that's the reason for the bean bags in John's photo above …
Offer students to pay in kind. Another social entrepreneurship option, which will be well received among students who come to Kathmandu from the villages, can be to offer them to pay with their own services for accommodation, not in money. That's quite simple here, as it's not just a rental space but a business that can offer paid work. Managing all the components of this symbiotic business cluster needs some low to medium skill manual labor, and people living on-site are the most natural candidates for that. They can work, for (say) 8 hours a week as barista, janitor, coffee roaster, coffee packaging, IT admin, cook, kitchen help, or guard. Pitfalls to avoid are (1) it must not result in a two-tier "class hierarchy" of those who can pay and those who must serve, (2) it has to be manage legally safe and sound (no "black labor") while avoiding or automating away the bureaucracy involved for hiring people.
Agile bedroom management. Since many students prefer low costs over high comfort, and because it offers additional capacity around events and high demand times, bedrooms should be more managed like a hotel. So, inhabitants and guests alike would not have their own permanent room, but get a free room when they come in. When they travel or sleep out, even if just for a night, they can (or have to?) clear the room, and in exchange don't have to pay for that time. All of this can be managed with open source resource scheduling software. For added comfort, as a permanent place to store their belongings, everyone would get a special rollable shelf that can even go up and down stairs like a sack barrow. To store more, the less used items can go into a personal, lockable storage cabinet in a special storage room or in the corridor. To accommodate even more guests during brief "peak demand" times, there could be additional beds in store, so that people comfortable with each other can share rooms.
Crowd-funding with rewards. To have the funds for furnishing and equipping the space (in a small initial version) including coffee processing equipment, a crowdfunding campaign seems very realistic. It would target mostly travellers from abroad who love Nepal and come back often. They can easily afford investing 100-500 USD each, in return for a guestroom they can book for free when in Nepal. This is aligned to member benefits from the proposed network of OpenVillage houses. 50 people donating 150 USD on average would yield 7500 USD, which is enough to start a small version of this space in Nepal. It can then grow from reinvesting its profits, or utilize growing popularity for follow-up "round B, C, D" crowdfunding campaigns with the same principle.
Crowd-investing with rewards. Another way to distribute crowdfunding rewards is transform it into a crowd-investing model, with with capped profits rather than permanent shares. For that, you create PayCoupons coupons for the space, to be redeemed against all products and services of the space after a certain holding period. For investing 100 USD, funders would get a for example 200 USD or 250 USD of coupons, and could either redeem them by themselves when in Nepal, or exchange them on PayCoupons for other coupons they want.
Sharing equipment. A good part of what makes co-living a low-cost alternative is sharing space. Similarly, by sharing equipment the rest of life can also cost less. For a co-living arrangement in Kathmandu, that would mostly be bicycles, electric bicycles, motorcycles / scooters, a small workshop with tools, perhaps a gym pass.
Own energy, own water, own food. A good part of the OpenVillage vision is self-supply, to lower costs of living. In Kathmandu, this is quite possible: (1) have a rainwater collection and filtering system, as that can provide enough water throughout the year, (2) add photovoltaics panels to the roof and have a 24 V DC system with batteries, for lighting and computers, (3) use self-built electrical bicycles and electrical scooters for transport, charged at home from photovoltaics power, (4) collect organic trash from the neighborhood and create biogas for cooking (and for roasting the coffee), using a 1-4 biogas digesters on the roof, (5) create compost from biogas digestate and other organic trash, (6) partially grow own food by rooftop and courtyard gardening. Neighbours could be persuaded to provide their rooftops for free if these gardens are nice to look at and to relax in the shadow of the plants. The equipment for all this technology should be mobile, prepared to move places without much effort. It would be standardized, resulting from a shared development effort among multiple OpenVillage houses.
