Pick an ecosystem, master it, then expand — not the other way around

The advice here is simple: pick an ecosystem you believe in for whatever reason, and become the best at that ecosystem. Once you've mastered it, if a client comes along and another model makes more sense for their situation, you can go recommend that. But your ability to determine what model is best for a task relies on your deep understanding of various models built over time through benchmarking. Don't try to jump to multi-model expertise before you've genuinely mastered one. Get good first, then venture out.