A ridge-beam can be supported by two tripods to make a pioneered "tripod-cruck" building. with the ends of tie-beams extended outward, the side walls can be framed just as in a conventional cruck building. This would work well. But while contemplating this idea I came across a picture I saved about six months ago that provided me with inspiration for how to raise a cruck building on uneven ground single handedly.
The alternative up to now was to build the crucks on the ground and raise them up like an old fashioned barn raising. Then brace them temporarily in place while I attach them to each other using some sort of scaffolding to allow me to work at the right height. Then brace the building and remove the scaffolding. That is a lot of work. If I built the crucks on the ground it would be less certain whether the feet of the crucks would be too long or too short, causing the tie-beam to be out of level. So this technique I am describing above sounds far more preferable. Without having to raise the crucks, I can use larger timbers too.
The crucks and tie-beams could be squared with an axe before tested for position, then marks could be made for lap joints where they cross. Timberlock screws could hold them in place, or large bolts (as Rob Roy has suggested).
This method for raising the tie-beam first before adding its supports could easily be used in the construction of a very simple trestle frame building. For a two-trestle building, use two tripods to level each tie-beam. Measure, cut, and wedge two posts (and concrete piers) under each tie-beam. Then place wall plates across the ends of the tie-beams. Everything should be plumb and square. Attach diagonal braces and remove the tripods. The roof could be simply framed with pre-made timber trusses, gusseted (or yoked) for strength, that are attached to the wall plates with bird's mouth notches. Or a ridge-beam could be supported by short posts, one centered on each tie-beam (not as strong, but simpler). To lengthen the building, add trestles and extend the wall plates centering joints over posts.
If a trestle is parallel to the slope of the hill, the post on one side will be longer than the post on the other. If it is perpendicular to the slope of the hill, posts will be equal lengths. If a series of trestles with tie-beams atop posts of equal lengths are arranged perpendicular to the slope of a hill, then the wall plates laid across them will have the same slope as the hill. In this way framing a roof may be unnecessary as the surface of the wall plates could form the shape of a shed roof with sufficient slope for shedding water when sheathed with roofing material. Just provide plenty of bracing. A word of caution: if the long axis of such a building is perpendicular to the slope of the hill, then the building could be composed of two very long trestles with multiple posts. But this reduces the strength of trestle frame construction, in which the shape of each trestle should not deviate too far from approximating a square and be composed of only two posts.
It goes without saying that logs are irregular, and these irregularities become compounded with each additional log. (The height of roofing material, rafter, wall plate, tie-beam, post, and foundation, added together, should be the same. Changes in the height of one need to be compensated for in another; ideally there is no need for this.) Therefore beginning the project with the straightest most consistent wood is best, otherwise be prepared to accept a lot of character in the final building. But character isn't bad, and construction techniques can be used to deal with irregularities, for example irregularities could be allowed to accumulate, then be compensated for at the end by varying the depth of the rafter's notch in the wall plate (assuming the wall plate is thick enough to absorb deep notches without compromising strength). Look at Ullandhaug Roman Age Farm in Stavanger, Norway. It has a lot of character and it is beautiful.
One of the biggest difficulties of building with round lumber is creating multiple joints whose surfaces all lie in the same plane. I think the key to overcoming this is to create a reference point or position against which they are all measured for accuracy. For example, rest a log securely on one surface (if it doesn't rest easily, attach it or create a base so that it can) then create all other cuts measured against that surface, whether plumb and vertical, level and horizontal, or at some other measured angle.
There are literally hundreds of measurements and cuts involved in timber frame construction. With dozens of cuts on each large member, any mistake is very costly as it could render the entire member useless. Stick frame construction, using only but joints, is therefore faster and easier by a wide margin. It would appear I have engaged in an exercise that has been solved long ago!