Extracting Textual Data using Machine Learning and Creating a Rules Engine

Project Description

We are looking for engineering help, under the guidance of our CTO, for the following work to support an ongoing project:

• Extracting specific textual data from content using machine-learning (TensorFlow and so on) such that the extraction gets better. A training data set can be provided. 
• Creating a rules engine for use with open source search software, such as Elastic Search, Lucene, or Solr. The rules engine will cause the search engine to override default search results and return specific data for queries.  The nature of these queries will include geolocation data, such as long/lat and zip code.
• Data scripting support, including some ETL moving flat files from staging to production in MySQL, data aggregation, and sheparding scripts to make sure they aren't failing (looking at log files, restarting as needed). Light DevOps stuff.
• Helping to automate data validation to ensure data moved by scripts matches source (using some level of fingerprinting a small subset of data and ensuring it matches after ETL). 
• System testing for quality assurance.

Ideally, this person or team will have experience with AWS, Python, search technology, machine-learning, and an interest in data fusion architectures - and doing interesting and fascinating work.