Was sanskrit first complex language

You are demonstrably wrong in your belief. Apart from the fact that your definition of "complexity" is far too simplistic and not subject to any kind of measurement, the question of language commensurability has long been settled in linguistics. Perhaps most famously, Sapir summarized the position thus in 1921:

When it comes to linguistic form, Plato walks with the Macedonian swineherd, Confucius with the head-hunting savage of Assam.

He's referring to the fact that all languages exhibit sufficient syntactic and morphological complexity to be put to any use a human mind can put them.

However, that does not necessarily mean that all languages will have developed the vocabulary to describe some of the issues that Plato or Confucius deal with. But that has little to do with complexity. Many of the languages that used to be called primitive address vast biological taxonomies, kinship structures with fine distinctions and subtle cosmogonies. All of these Plato or Confucius would find difficult to address. Also, we should not confuse specialised vocabulary (such as that used by theologians, nuclear physicists or chartered accountants) with complexity.

All this obviously applies historically as well synchronically. Therefore, not only was Sanskrit not the first 'complex' language, it was not anywhere near the stage where language first achieved the structure and complexity it has now. No language prior to about 6000 years ago has been recorded but the material culture in evidence suggest complex communication tens of thousands of years into the past. However, this brings us to questions of the evolution of language complexity in general which are far outside what can be addressed in this context.

There have been some attempts recently to slightly soften this strict commensurability stance but certainly not to the degree your question expects.