I've spent a similar amount of time as you developing an automated system, and am trading it live. Be extremely cautious about..... 1) Backtesting Results. Just because something has worked in the past doesn't mean it will be profitable in the future. I had great results from backtesting, but when I forward tested the profitability was significantly reduced. The markets are in a constant state of flux, and you need to understand when your system is profitable, and when it's not. 2) 10k start $. This is way way way not enough, even for a single ES contract. It is a certainty that the day you start using your system live, the markets will do a backflip and you'll get a big draw-down. 3) The e-mini. IMO, the e-mini is (by far) the most difficult contract to auto-trade, since it exhibits far more randomness than other contracts. Try the Russell 2000.
Beside all the very valuable remarks, I believe another very important aspect on an intrady stragy is that if you have backtested it on lets say 1 or 5 minute historical bars crossing some indicators (SMA, whatsoever) you would by far not see the same (good) fills in real trading. Indicators are (often) built on the close of a bar, which might be 4mins and 59sec after you entered the live trade (on 5min bars), so the moment you enter the signal is way off compared when you look at the chart afterwards. Well my explanantion is bad, but 1 or 2 ticks worse entry fill and 1 or 2 ticks worse exit probably will turn a nice looking ES based intraday system into a loser. Papertrade it first, and then trade it with the lowest possible quantity. If your backtest shows 5 consecutive periods as a worst case, prepare yourself to see 6 to 10 when starting. 10k and 2 ES would not be my thing. 100k and 2 ES would make me feel better, specially for the beginnig.