I'm looking for any advice on backtesting/forward testing engines with ML framework. There seems to be a lot of choice when it comes to ML, although not all are geared towards trading (eg. https://github.com/GoogleCloudPlatform/cloudml-samples).
That's because the problem of ML is not straight forward or easy in finance. Pick up Advances in Financial Machine Learning to get an idea. I'm not turning you off to the idea but it's not easy and you should appreciate the difficulty before going down this rabbit hole. NeuroShell trader or whatever is the only one I can think of. Otherwise, you're on your own. Python + Scikit Learn + Pyalgotrade or something is probably where you're gonna want to go.