Some kinds of work are so tricky or complicated that practitioners can't guarantee results; they can only promise "best efforts" and (perhaps) "best practices". I'm not thinking here about work like farming or fishing, where the unpredictability of nature is a large part of what happens. Rather, I'm concerned with tasks whose character is inherently tricky or unpredictable.
So technical work is work that requires a fair amount of expertise to carry out, and whose results are inherently unpredictable.
Every occupation involves some technical work, because dealing with people is technical work, and every occupation deals with people. Some occupations have technical work at their core: the traditional learned professions (medicine, law, and clergy), military officers, the fine and performing arts, the various kinds of engineering, architecture, science. Many of traditional crafts are also technical work at the core: making musical instruments, for example. And some occupations often deal with people's identities in fundamental ways, even though they aren't considered professions; much of police work is like this, for example.
My focus is on the tasks, not the people who carry them out or the occupational categories. Keeping these things separate lets us frame a whole series of interesting questions. For example: how do tasks get assigned to occupations, and what sorts of issues arise as a result? The boundaries among academic disciplines, for example, is one phenomenon which I'm particularly concerned with. Boundaries among professions is another example; Andrew Abbott's book on the subject is especially interesting.