As laid out here, I talk about becoming a staff scientist in a company. ***Follow up, I talk about some presentation skills here that matter for staff scientists****
This implies you have done a post doc (eithe acadmic or corporate) and have some product development or drug development experience under your belt. You have been in the corporate world for awhile and are, to some degree, ready to lead a team of some size.
When you boil this all down, you are just a PI like the academic world, except you are making better money. You don't write grants, but you do write reports. You don't worry about tenure but you do worry about how well the company is doing.
You can get laid off more easily than a professor can. It also happens more frequently than most universities seem to prune their staffs. This is, to a great degree, dependent on the size of the company you are at. Larger companies lay off less frequently, but when they do so it is a VERY big deal. Smaller companies, jobs are less stable at, but the potential rewards of stock options are there. Your personality will determine which you want to work at.
The job itself is really much the same. You have a project to do, and you have to work to try and get it done. the fact that the project may have to conform to corporate goals is, from my point of view, no different than the fact that you have to conform to what you wrote in your grant. People always talk of the "academic freedom", but I have always been an un-beleiver. You have, to a great degree, to follow what the subject of your grant was or you will have problems with continued funding. Attempts to make serious turns (in the academic world) are met with funding problems. For everything, there are exceptions, but for the most part I think this a true statement.
The Staff Scientist will plug away at the project. After some years, you may move up to where you have multiple groups reporting to you. Now you are setting higher level research priorities, and the groups will be plugging away. If you get high enough up, you will be picking which diseases to work on, and will be coordinating groups across mutiple sites (and likekly multiple continents).
I think a major difference between academia and the industry is that, at many companies, it is actually a LOT more collaborative than in academia. I have had a good look at Pfizer, and there all the groups that work on diseases do actually share a lot of information. There are, I am sure, hold outs, but they are WAY more sharing than similar academic groups would be. This enables you to move a LOT more quickly than academic groups as you have, to the degree they can manage, less repetition of work. The company wants to do things as few times as it can in order to get that knowledge, so the sharing is deeply encouraged.
I am, obviously becuase of my choice, biased toward the industry side. I can say that at my company we have the world experts on this one thing (won't say what, as that would peg us). There is no one, anywhere, that knows more than our group. I am pretty sure of this becuase I spend a great deal of my time off looking for any threats, and then will attempt to buy those threats. The area comprises very basic research, but with the intent that the full understanding of that area allows us to ship cutting edge products in that area. You have to be familiar with every amino acid and the function and shape of it in order to be able to put out the next generation that is faster/better/cheaper. That means we do a lot of what most people would call 'basic' research. It also means that we publish in patent applications, and then are amazed when the academic world gets a nature paper for data that is repeating what is in published patent applications....Academics should read a wider scope of article!