Update docs

Author: deliahu

README.md

@@ -7,9 +7,9 @@
 
 <br>
 
-**Machine learning infrastructure for developers: build  machine learning applications faster.**
+**Machine learning infrastructure for developers: build machine learning applications faster.**
 
-Cortex is an open-source platform for building, deploying, and managing machine learning applications in production. It is designed for any developer who wants to build machine learning powered services.
+Cortex is an open-source platform for building, deploying, and managing machine learning applications in production. It is designed for any developer who wants to build machine learning powered services without having to worry about infrastructure challenges like configuring data pipelines, continuous deployment, and dependency management.
 
 Cortex is actively maintained by Cortex Labs. We're a venture-backed team of infrastructure engineers and [we're hiring](https://angel.co/cortex-labs-inc/jobs).
 
@@ -19,9 +19,9 @@ Cortex is actively maintained by Cortex Labs. We're a venture-backed team of inf
 
 - **End-to-end machine learning workflow:** Cortex spans the machine learning workflow from raw data ingestion to prediction serving.
 
-- **Declarative configuration:** Cortex applications are defined using a simple and modular syntax that enables flexibility and reusability.
+- **Machine learning pipelines as code:** Cortex applications are defined using a simple declarative syntax that enables flexibility and reusability.
 
-- **Scalability:** Cortex is designed to scale to datasets of any size.
+- **Scalability:** Cortex is designed to scale to handle production workloads.
 
 - **Spark & TensorFlow support:** Cortex supports custom [Spark](https://spark.apache.org/) code for data processing and custom [TensorFlow](https://www.tensorflow.org/) code for model training.
 

docs/applications/resources/environments.md

@@ -1,6 +1,6 @@
 # Environments
 
-Transfer data at scale from data warehouses like S3 into the Cortex cluster. Once data is ingested, it’s lifecycle is fully managed by Cortex.
+Transfer data at scale from data warehouses like S3 into the Cortex environment. Once data is ingested, it’s lifecycle is fully managed by Cortex.
 
 ## Config
 

docs/applications/resources/overview.md

@@ -1,6 +1,6 @@
 # Overview
 
-Cortex applications consist of declarative resource configuration written in YAML as well as Python code to implement aggregators, transformers, and models. Each resource has one of the following `kind`s:
+Cortex applications consist of declarative resource configuration written in YAML as well as Python code to implement aggregators, transformers, and models. Each resource has a `kind`:
 
 * [app](app.md)
 * [environment](environments.md)

docs/quick-start.md

@@ -21,7 +21,7 @@ Jump to [deploy the application](#deploy-the-application).
 
 ## Build a machine learning application
 
-Let's build and deploy a classifier using the famous [Iris Data Set](https://archive.ics.uci.edu/ml/datasets/iris)! Below are a few samples of iris data:
+Let's build and deploy a classifier using the famous [iris data set](https://archive.ics.uci.edu/ml/datasets/iris)! Below are a few samples of iris data:
 
 |sepal_length|sepal_width|petal_length|petal_width|class|
 |:---:|:---:|:---:|:---:|:---|
@@ -71,7 +71,7 @@ Cortex will be able to read from any S3 bucket that your AWS credentials grant a
 
 #### Define raw columns
 
-The Iris Data Set consists of four attributes and a label. We ensure that the data matches the types we expect, the numerical data is within a reasonable range, and the class labels are within the set of expected labels.
+The Iris Data Set consists of four attributes and a label. We ensure that the data matches the types we expect, the numerical data is within a reasonable range, and the class labels are within the set of expected strings.
 
 Add to `app.yaml`:
 
@@ -183,7 +183,7 @@ Add to `app.yaml`:
 
 #### Define transformed columns
 
-Transformers convert the raw columns into the appropriate inputs for a TensorFlow model. Here we use the built-in `normalize` and `index_string` transformers using the aggregates we computed earlier.
+Transformers convert the raw columns into the appropriate inputs for a TensorFlow estimator. Here we use the built-in `normalize` and `index_string` transformers using the aggregates we computed earlier.
 
 Add to `app.yaml`:
 
@@ -300,7 +300,7 @@ Cortex supports any TensorFlow code that adheres to the [tf.estimator API](https
 
 #### Define web APIs
 
-This will make the model available as a live web service that can make real-time predictions.
+This will make the model available as a live web service that can serve real-time predictions.
 
 Add to `app.yaml`:
 
@@ -538,7 +538,7 @@ Define a sample in `irises.json`:
 }
 ```
 
-When the API is ready, run the prediction like so:
+When the API is ready, request a prediction from the API:
 
 ```
 $ cortex predict iris-type irises.json