The EVA Data Integration Project addresses actions from a 2014 EVA on-orbit incident and long standing challenges in gaining accurate, efficient, and effective use and access to EVA data and EVA data systems. This project supports the EVA Management Office by designing, developing, and delivering a new set of data management services, and developing information architectures to organize and create useful forms of EVA data for current and future work. The objective is to support the EVA Management Office’s responsibility that all EVA data and documentation is accessible, managed, controlled, and archived to reduce the effort of finding, using, and managing EVA data through the introduction of various integrated capabilities, such as data browsing, searching, integration, and review.
We are seeking a data analytics professional to assist the analysis of this data and help inform the types of analytics tools to incorporate in the EVA data system. In this role the successful candidate will:
Work with simulated data record (both historical and real-time) and create ways to tag the data, generate metadata using new metadata standards (e.g. hdf5) and parse the data into usable formats using aggregation and filtering techniques
Create a queuing strategy to will handle asynchronous call to data from source to location. The user will be agnostic about performance and/or query. (Distributed logger with distributed high performance pub-sub messaging using Apache Kafka)
Create new document-based database schemas (json-based) that can be deployed in a distributed manner to solve latency and bottlenecks for large queries. (data sinks using systems like rocksDB/Cassandra)
Create and deliver prototype that will demonstrate the feasibility of using open source packages to deal with streaming data (e.g. real time data validation using apache spark)
Create both hard and soft real-time analytical modules based on machine learning algorithms to extract useful features from data. The hard module can be an embedded GPU board (e.g. Nvidia Jetson) that runs a forward trained model to analyze patterns and behavior in data streams. The soft module is more of an offline analysis tool for generating recommendations and alerts.
Deliver an onboard visualization platform for fast exploratory analysis.
Review and recommend technologies and architecture design
Design and recommend system specifications that can be used to deploy the data analytics platform. Requirements
A degree in Computer Science or related discipline with emphasis on data analytics
Experience with Database Systems, Relational and NoSQL
Experience analyzing big data with metadata tagging
Data Interchange Formats: XML, JSON
Experience developing or integrating data visualization capabilities
Knowledge of implementing pub/sub capabilities, especially with Apache Kafka
Exceptional written and oral communication skills
You are submitting your information for consideration to a position that is currently available under the Intelligent Systems Research & Development Services 2 (ISRDS-2) program at Ames Research Center, CA. If selected, your employment offer may be with SGT or one of our teammates on the contract.
To apply for this job please visit tinyurl.com.