(resume in 1-page PDF file)

Andrew Ganse, PhD
andrew@ganse.org | http://research.ganse.org | https://www.linkedin.com/in/andrew-ganse


PROFESSIONAL EXPERIENCE

Senior Imaging Data Scientist, Thruwave Inc. July 2020 – present.
• Owned and grew the company’s data science capability from scratch, enabling its 3D millimeter wave radar imaging systems and RGB/depth cameras to deliver actionable results at scale to customers for fraud, warehouse inventory management, quality control (e.g. divert sealed boxes with fraudulent contents).
• Successfully completed 10+ pilots at major corporations and agencies, resulting in multiple conversions to revenue-generating customers.
• Leading a 4-member data science team in building out the system and conducting customer analyses, iterating directly with product manager, CTO, and CEO to best fit technical plans to business goals.
• Designing computer vision and machine learning for classification, boxbounds, and anomaly detection; OpenCV, Scikit-learn, Keras/Tensorflow, SciPy, Python/Pandas, PostgreSQL, Docker, MLflow, AWS.
• Patent: Ganse, A.A., et al. (2022). Systems and methods for high throughput anomaly detection with 3D millimeter wave imaging. U.S. Patent No. US20240192148A1, filed 2022-Mar-31. Patent pending.

Radar Data Scientist, R&D Team, Echodyne Corporation. Dec 2017 – July 2020
• Owned and designed the company’s field-test database and tracking performance system for its novel meta-materials radar, supplying performance specs for its marketing literature, resulting in customers responding: “This company is the only one whose radars do exactly what their literature says they do!”
• Led a team to build that evaluation system and use it in operations, iterating with product managers to create 10s of analysis reports from 100s of field-tests, resulting in ongoing customer conversions.
• Initiated and co-developed the company’s first tracked-object classifier, on its EchoGuard product, a machine learning feature which quickly became one of the most requested by customers for that product.
• Led a team to research and develop machine learning methods for automotive radar tracking and classification, integrating them into the cognitive radar system for the company’s AV test vehicle.
• Technical management and mentorship of 2-5 people per project in multiple concurrent projects; co-organizer of the company’s intern program.
• Python/Pandas, PostgresSQL, Docker, RabbitMQ, MLflow, Scikit-learn, Keras/Tensorflow, Nvidia GPUs.

Principal Scientist, Anseres Research & Technology LLC. Sept 2016 – Dec 2018.
• Scientific R&D consulting in defense and space science; completed multiple federal subcontracts.
• Published radio-science gravimetry research at conference with NASA & university collaborators.
• Led SBIR proposal submission on Deep Learning for Clutter Reduction in [Sonar Systems], with university collaborators.
• Internal R&D predicting estuarine salinity from electric fields using TensorFlow neural networks.

Data Scientist, Spare5. Jan 2016 – June 2016.
• Developed machine learning models for data quality and user reputation evaluation on Spare5’s intelligent-crowdsourcing platform for data labeling. Word2vec, PostgreSQL, R/Rserve, Python/Pandas.
• Produced model that resulted in 8% lift in search-traffic to a customer’s data in first few months of job.
• Engaged in press/analyst briefings, meetings with customers, public speaking (Datapalooza).

Senior Research Physicist, Applied Physics Laboratory, Univ. of WA. Apr 1999 – Nov 2015.
• Designed research experiments, conducted applied physics research in inverse problems and wave propagation in random media, in seismo-acoustic, electromagnetic, and gravimetric domains.
• Published research results in 25 publications (including conference publications).
• Solved nonlinear regression, inversion, optimization, tracking, and signal processing problems in acoustic, seismic, electromagnetic, and gravity remote sensing applications; using Python, Matlab, Octave, C, Java, Fortran, Linux shell scripting; administrated Linux clusters.
• Developed Kalman nonlinear smoothing/tracking and parameter estimation algorithms to acoustically track a 5km(!) long vertical hydrophone array for our ocean acoustic experiment.
• Analyzed fluctuations in intensity and pulse spreading of ocean acoustic signals interacting with ocean internal waves, testing prevailing theory with our at-sea experimental measurements.
• Created Markov Chain Monte Carlo based Bayesian inversion of acoustic data on a Linux cluster.
• Organized technology transfer of a new APL-developed technology to a small business.
• Designed research experiments, presented results at conferences, wrote proposals and reports to sponsors, interfaced with sponsors, wrote and reviewed research papers.
• Led system engineering and field testing for an experimental “towed CTD chain” cabled instrument containing 90 sensors, managed teams of 2-6 others at a time in testing and operation of the system. Managed two students in statistical programming projects.

Database Programmer, Pacific States Marine Fisheries Commission. Jan–Dec 1998.
c/o National Oceanic & Atmospheric Administration (NOAA) National Marine Fisheries Service.
• Programmed data processing and extraction routines for an Oracle database of commercial fishing in western U.S.
• Provided expertise in Oracle, SQL, UNIX to two PSMFC agencies.


EDUCATION

Ph.D. Geophysics, University of Washington.
B.S. Electrical Engineering, University of Washington.


SELECTED SIDE PROJECTS (see GitHub account and this website)

Flow_models: Flow-based invertible neural networks via TFP for generative image modeling.
Gpt_client: CLI-based OpenAI/ChatGPT client with model selection, token tracking, and URL parsing.


ADDITIONAL EXPERIENCE

Industry consultant for undergraduate computer vision project. Jan 2019 – Apr 2019.
DigiPen Institute of Technology; Prof. Jeremy Thomas, co-advisor.
Advised student on senior project in computer vision for traffic flow analysis based on automated photo recordings, using a pre-trained deep learning model.

Recurring guest lecturer, graduate-level inverse theory (statistical inference). 2008 – 2015.
University of Washington, Earth and Space Sciences Department.
Wrote and taught 1.5-hour lectures for numerous graduate-level geophysical inverse theory classes of ~10 students. Developed seven computational lab assignments.


RECENT PUBLICATIONS AND PRESENTATIONS

• Ganse, A.A., C.M. Madden, C.M. Watts, A. Pedross-Engel, M.S. Reynolds, "High-throughput anomaly detection in document envelopes with 3D millimeter wave imaging". Proc. SPIE, 11745-14. Passive and Active Millimeter-Wave Imaging XXIV. April (2021).
• Ganse, A.A., “An eigenspectrum filter-factors approach to interpreting regularization and subspace methods”, presentation at Echodyne Corporation, Dec 2019.
• Roberts, J.H., S. Vance, A.A. Ganse, “Detection of Gravity Anomalies on Europa using Line-of- sight Gravity Profiles”, Abstract P42B-06, Fall Meeting AGU, San Francisco (2018).
• Andrew, R.K., A.A. Ganse, A.W. White, J.A. Mercer, M.A. Dzieciuch, P.F. Worcester, “Low- frequency Pulse Propagation over 510 km in the Philippine Sea: A Comparison of Observed and Theoretical Pulse Spreading”, J. Acous. Soc. Am., 140, 1 (2016).
• Ganse A.A., S. Vance, and J. Roberts (2014), "Inverse theory resolution analysis in planning radio science gravity investigations of icy moons", Abstract P43C-3997, Fall Meeting AGU 15-19 Dec, San Francisco (2014).
• Ganse A.A., R.K. Andrew, F.S. Henyey, J.A. Mercer, P.F. Worcester, M.A. Dzieciuch, “Model and data comparisons of ocean acoustic intensity statistics in the Philippine Sea 2010 experiment”, J. Acoust. Soc. Am. 135 , 2306 (2014).


FIELD EXPERIENCE

• 2010 : Philippine Sea long-range ocean acoustic experiment cruise (UNOLS R/V Revelle), 5 weeks.
• 2009 : Philippine Sea long-range ocean acoustic engineering test cruise (UNOLS R/V Melville), 5 weeks.