Think speed could be a potential benefit. Why go through source -> llvm ir -> binary with linking and all the other stuff? Think I’m bearish on the concept, but who knows?
One parallel could be using ml for simulations to not write or compute all the rules
Comments
If true -- why? What would be the advantage?
Think speed could be a potential benefit. Why go through source -> llvm ir -> binary with linking and all the other stuff? Think I’m bearish on the concept, but who knows?
One parallel could be using ml for simulations to not write or compute all the rules