Ok found out how. Indexing on MultiIndex

multiple_df.index.name = None
data = {idx: gp.T for idx, gp in multiple_df.T.groupby(level=0)}

separated_df = {}
for ticker in data:
    new_df = data[ticker][ticker]['Close High Low Open Volume'.split()]
    new_df.columns.name = None
    adj_close = data[ticker][ticker]['Adj Close']
    new_df.insert(0, 'Adj Close', adj_close)
    
    separated_df[ticker] = new_df

print(separated_df)