Download PDF

Sang Han

Machine Learning Engineer


Participated as team lead on MEMEX, a DARPA project out of the  Information Innovation Office to develop next generation search and extraction systems. The initial application is the fight against human trafficking.

Created state of the art deep learning and factor networks based inferencing techniques resulting in over $4M in DoD funding released to Qadium towards developing our core technologies. Currently, every human trafficking investigation pursued by the Human Trafficking Response Unity in New York City involves MEMEX.

Work History


Machine Learning Engineer


At UnifyID we're trying to solve the problem of authentication. Using factors like gait and location, we are able to authenticate individuals into their accounts.

  • Built a data pipeline for ingesting and processing raw sensor data at a terabyte scale.
  • Optimized training from single machine Jupyter notebooks to distributed training over GPU clusters.
  • Researched novel methodologies for spatiotemporal segmentation using quaternion based representative geometry.

Data Scientist

Qadium, Inc

Qadium creates and organizes knowledge about the world's devices. As a Data Scientist at Qadium's early stage, I helped build out the software for several federally awarded DARPA contracts.

  • Authored several software projects developing new graph based entity search technologies for deep and adversarial domains available on DARPA open catalog.
  • Open sourced a python library automating internet scale text extraction and data enrichment for EMR jobs running over common crawl.

Junior Data Scientist


Blockspring is an API programming platform. As a Junior Data Scientist, I helped train initial machine learning models that would eventually go into their first NLP "Blocks" api.

  • Worked under Peyman Faratin in building NLP stack
  • Trained Named Entity Recognizer based on Stanford CoreNER optimized for BLEU scoring

MEMS Process and Data Engineer

Calient Technologies Inc

Precision engineering of microelectromechanical systems to specifications of single microns, from virgin silicon wafers through its final packaging and testing.

  • Acquired experience in seeing through a complex engineering project from start to finish.
  • Used k-means to cluster bins of silicon dies



awards and distinctions