Data Scientist - Fabrication Facility 15, Saijo, Hiroshima, Japan

Role & Responsibility

  • Draw from a broad background of data-minning technique in mathematics, statstics, information technology, machine learning, deep learning and AI, data engineering and experiements, visualisation etc. to discover insightfull patterns in seminconductor manufcaturing data.
  • Design and implement optimum data strctures in the appropriate data managment system (Hadoop, Hive, Hbase, Teradata, SQL Server etc.) to satisfy the data requriements.
  • Define data-qality objective for the solution.
  • Devlope processes to efficiently load and traform the data into the data management system using SQL, PL/SQL, PySpark, Shell Scripting etc.
  • Developing new or enhancing prior data acquisiation and ETL pipeline from various sources into bid data ecosystem.
  • Developing expertise in data mining and analytic methods.
  • Determine statistical validty and significance (pick out signals and noise).
  • Idenitfy and aplu appropiate algorithm or analytical model
  • Develop predictive models.

Projects Done

  • Improving Micron Memory Yield Efficiency, Foundries fabrication equipment data integration with other sites. Using it to optimize Yield Efficiency.
    • Tools Used: Spark, Pyspark, python, Hive, Tensorflow, Spark-sklearn.
  • Testing of "Deep Learning + Big (NoSql + Structured)Data Base + Distributed Computing Algorithms"
    • Tools Used: Tools Used: H2O.Ai, Sparkling Water, TensorFlow, Intel BigDl.