diff --git a/README.md b/README.md
index b0bc619..9ed878b 100644
--- a/README.md
+++ b/README.md
@@ -21,8 +21,119 @@ Project examples found under [tensorflow-ue4-examples](https://github.com/getnam
 
 It is also the main repository where all development is tracked for all plugin dependencies for tensorflow.
 
+## Python API
+
+You can either train directly or use a trained model inside UE4.
+
+To start, add your python script file to _{Project Root Folder}/Content/Scripts_.
+
+wrap your tensorflow python code inside this API:
+
+
+#### MySubClass(TFPluginAPI)
+
+import the following and subclass in your module file
+
+```python
+import tensorflow as tf
+import unreal_engine as ue
+from TFPluginAPI import TFPluginAPI
+```
+
+Add the following class functions
+
+```python
+class ExampleAPI(TFPluginAPI):
+
+	#expected api: setup your model for training and any member storage info
+	def setup(self):
+		pass
+	#expected api: using the stored session and class data, evaluate the json inputs
+	def runJsonInput(self, jsonInput):
+		result = {}
+		return result
+
+	#expected api: start training your network
+	def train(self):
+		pass
+    
+#NOTE: this is a module function, not a class function. Change your CLASSNAME to reflect your class
+#required function to get our api
+def getApi():
+	#return CLASSNAME.getInstance()
+	return ExampleAPI.getInstance()
+```
+
+Note the ```getApi()``` module function which needs to return a matching instance of your defined class. The rest of the functionality depends on what API you wish to use for your use case. At this stage all input/output from UE4 is JSON encoded.
+
+If you have a trained model, simply setup your model/load it from memory and omit the training function.
+
+If you wish to train ensure you check for ```self.shouldstop``` after each batch/epoch to handle early exit requests from the user.
+
+A slightly more expanded example api:
+
+```python
+class ExampleAPI(TFPluginAPI):
+
+	#expected api: setup your model for training
+	def setup(self):
+		#setup or load your model and pass it into stored
+		
+		#Usually store session, graph, and model if using keras
+		self.sess = tf.InteractiveSession()
+		self.graph = tf.get_default_graph()
+
+	#expected api: storedModel and session, json inputs
+	def runJsonInput(self, jsonInput):
+		#e.g. our json input could be a pixel array
+		#pixelarray = jsonInput['pixels']
+
+		#run input on your graph
+		#e.g. sess.run(model['y'], feed_dict)
+		# where y is your result graph and feed_dict is {x:[input]}
+
+		#...
+
+		#return a json you will parse e.g. a prediction
+		result = {}
+		result['prediction'] = -1
+
+		return result
+
+	#expected api: no params forwarded for training? TBC
+	def train(self):
+		#train here
+
+		#...
+
+		#inside your training loop check if we should stop early
+		#if(this.shouldstop):
+		#	break
+		pass
+
+#required function to get our api
+def getApi():
+	#return CLASSNAME.getInstance()
+	return MnistSimple.getInstance()
+
+```
+
+a full example can be seen here: https://github.com/getnamo/tensorflow-ue4-examples/blob/master/Content/Scripts/mnistSimple.py
+
+### Load your module from your TensorflowComponent
+Select your TensorflowComponent
+
+![select component](http://i.imgur.com/f9Syql1.png)
+
+and change the TensorFlowModule to reflect your _filename_ without .py.
+
+![change module name](http://i.imgur.com/mpzymgd.png)
+
 
 ## [License](https://github.com/getnamo/tensorflow-ue4/blob/master/LICENSE)
 Plugin - [MIT](https://opensource.org/licenses/MIT)
 
 TensorFlow and TensorFlow Icon - [Apache 2.0](http://www.apache.org/licenses/LICENSE-2.0)
+
+
+