Ask HN: Which languages and frameworks do frontier LLMs understand best?

I'm exploring ways to maximize my productivity by leveraging the capabilities of frontier LLMs to generate most of my code. Given the rapid advancements in AI, I'm curious about which programming languages and frameworks these models currently understand and handle most effectively.

Specifically, I'm debating whether to focus on TypeScript + React or stick with Python. Are there advantages in how LLMs perform with one over the other in terms of code generation, debugging, or providing more accurate suggestions? Additionally, any insights into how certain tools or ecosystems might enhance the interaction with these models would be greatly appreciated. Looking forward to hearing your thoughts and experiences!