Case-specific extended attributes for employee dimension

I have a dimension "Employee" which has a very standard set of attributes. The company the employees work for have many clients and employees work on-site for a given client.

When an employee works on-site for an end-client, they have attributes specific to that client.

Theoretical:
Todd works FOR the company Staffing Company but physically works AT Widget Company.

In this case, Todd has a whole extended set of attributes that are specific to JUST Widget Company:
* contractor_id
* ldap_login
* job_title (which is different than Todd's job title at Staffing Company
* etc.
* etc.

How do you handle this situation dimensionally? Imagine there are literally over 1,000+ different types of Widget Company's which may very well have their own extended attributes we need to hold on to.

Any insight would be appreciated.

Thanks!

An employee working at a site is a business event. Events are represented by fact tables. The employee, client, title... are all contexts of that event.

Thank you so much for your response. I can understand your suggestion, particularly if I was trying to model sparse / varying placements of an employee, however, in my situation an employee is placed permanently in that location, sometimes for years at a time. At the end of that time, the employee typically ceases to be an employee because they are no longer needed. So, it's almost as if these extended attributes apply to the employee as a whole vs just for an event (again, different attributes when the employee is assigned a different placement location).

Would this still be modeled the same way given the fact that the employee has the same placement sometimes for years and it never changes?

As always, thank you for your help.

Yes. It is the basic dimensional form. There are a lot of ways you can model it, but if you are doing a dimensional model, that is the way it is done... a fact table with dimensions.