Check out page 46 of the Valve handbook, I think their take offers some insight: http://www.valvesoftware.com/company/Valve_Handbook_LowRes.pdf http://www.valv

Check out page 46 of the Valve handbook, I think their take offers some insight: http://www.valvesoftware.com/company/Valve_Handbook_LowRes.pdf http://www.valvesoftware.com/company/Valve_Handbook_LowRes.p... On the other hand, there's an argument to be made to the tune of: "Try to sit in two chairs and you'll fall between them." My opinion is that anyone who builds and/or leads a successful startup will tend to wear many hats during the process, especially in the beginning, so if you're goal is to found a startup, having a diverse skill set can't hurt. Even if the founding team consists of individuals with their own unique skills, the cycle of design, production, and deployment is one that requires a collaboration between stages to be functional. Design leads to production, production to deployment, and deployment to re-assessment and design once again, etc. Stagnation in any of those areas is dangerous for the success of a startup, and can come from a missing or weak link, and just as easily due to poor transitions between phases. Being able support this cycle by tapping into both skills is a great asset to a team, and a early hire or a founder, and I believe that creates a strong argument for maintaining each skill set.