Would there be any benefit to using reactive programming techniques with a severless framework? As I understand it lambda will spin up new instances if all existing instances are processing a request. if an already spun up instance was waiting on a db but made good use of reactive/concurrent programming would this instance then appear as "free" to handle a request or would it make no difference to lambda?

There would be no difference. Basically lambdas are already parallelization technique and any reactivity should be build upon them

that is what i thought and couldnt find much info on it.