Conjecture :
$$\prod_{n=1}^{\infty}(1+e^{-n})=^{?}1/1!+1/2!+1/3!+0/4!+1/5!+1/6!+7/7!+5/8!+9/9!+7/10!+\cdots+a_n/n!+\cdots$$
Where $a_n$ is an integer such that :
$$0\leq a_n\leq n$$
Some arguments :
I think the best we can say is until we have the (says 20 because of floating point ) first term which seems true currently more it goes more we have choice in the $a_n$ integers more the conjecture shall be true .
Another argument : if we stop to $a_k$ and take all the next terms equal to $n$ it seems always greater than the product .
Context :
For a context I was using the Engels form of $\ln 2$ which is near from the value of the product . There is also the Jacobi theta function as start for a formal proof .
Question :
How to (dis)prove the conjecture ?
Second question :
Following the comment what is the factoradic representation starting from $1!$ ?