Do you need to hire somebody for this? It frankly doesn't seem like a hideously difficult problem and probably doesn't require a backend of any sort, except maybe to store logs for developing better recognition.
I'm not sure if you're trolling me, but looking through your other comments seems like you are a serious commenter :), so I'll treat your query as bona fide.
The main problems are:
* transmission of data in a reliable fashion (both from the phone and the module which plugs into your car) to the decision system
* making decisions quickly and accurately based on incomplete data (due to dead zones, time shifts, etc)
* displaying decisions (building a UI both for the end user and for administration)
* integrating with telcos (since it is a carrier level solution)
Now, this problem isn't the same level of difficulty as say, "scaling AWS" or "inventing Google Maps", but I've found it to be an interesting problem. It certainly is more than a phone only solution, which you imply.
That said, maybe I'm missing something obvious that would make the solution simple--I'd be happy talk to the hiring manager about paying you for an hour of your time to chat about your proposed solution if you'd be interested. My contact info is in my profile, so feel free to ping me with your rate and availability.
* transmission of data in a reliable fashion (both from the phone and the module which plugs into your car) to the decision system
* making decisions quickly and accurately based on incomplete data (due to dead zones, time shifts, etc)
* displaying decisions (building a UI both for the end user and for administration)
* integrating with telcos (since it is a carrier level solution)
There is no way to know whether someone is driver or passenger without them volunteering that info. If they are willing to tap a button that says they are driving they could just easily press a button that disables talk/text.
