David M. Andrzejewski


RECENT PROFESSIONAL EXPERIENCE

Industry

Sumo Logic
Director of Engineering, AI Experiences
Redwood City, CA (2019-present)

Sumo Logic
Various engineering roles
Mountain View and Redwood City, CA (2011-2019)

Lawrence Livermore National Laboratory
Postdoctoral Research Staff Member
Livermore, CA (2010-2011)
-Apply statistical modeling to knowledge discovery in text corpora

Academic

University of Wisconsin-Madison
Research Assistant (Professors Mark Craven and Xiaojin Zhu)
Madison, WI (2008-2010)
-Project: Knowledge-augmented topic models
-Developed new latent topic models to allow prior knowledge and user feedback
-Proposed, implemented, and conducted experiments on new models and techniques


EDUCATION

University of Wisconsin-Madison
-PhD, Computer Sciences 2010
Research focus: Machine learning
Advisors: Mark Craven and Xiaojin Zhu
Thesis: Incorporating Domain Knowledge in Latent Topic Models
-MS, Computer Sciences 2007
-BS, Computer Engineering, Mathematics, Computer Sciences 2005


SELECTED PUBLICATIONS

David Andrzejewski and David Buttler. Latent topic feedback for
information retrieval. In KDD '11: Proceedings of the 17th ACM SIGKDD
Conference on Knowledge Discovery and Data Mining, 2011. Association
for Computing Machinery. (8% of submissions accepted for oral
presentation)

David Andrzejewski, Xiaojin Zhu, Mark Craven, and Benjamin Recht. A
framework for incorporating general domain knowledge into latent
Dirichlet allocation using first-order logic. In IJCAI ’11:
Proceedings of the 22nd International Joint Conference on Artificial
Intelligence, 2011. AAAI Press. (17% of submissions accepted)

David Andrzejewski, Xiaojin Zhu, and Mark Craven. Incorporating domain
knowledge into topic modeling via Dirichlet forest priors. In ICML
'09: Proceedings of the 26th Annual International Conference on
Machine Learning, pages 25-32, 2009. Association for Computing
Machinery.  (25% of submissions accepted)

David Andrzejewski, Anne Mulhern, Ben Liblit, and Xiaojin
Zhu. Statistical debugging using latent topic models. In ECML '07:
Proceedings of the 18th European conference on Machine Learning, pages
6-17, 2007. Springer-Verlag.  (9% of submissions accepted).

SELECTED TECHNICAL TALKS

TALKS

Reliable machine learning. Scale By the Bay, Oakland (November 2019)

Privacy-aware data science in Scala with monads and type level
programming. Scale By the Bay, San Francisco (November 2018)

Understanding Software System Behavior With ML and Time Series
Data. QCon.ai, San Francisco (April 2018)

Functional Programming for ML (panel). Scale By the Bay,
San Francisco (November 2017)

Economical ML via functional programming. 
Big Data Scala by the Bay, Oakland (August 2015)

Graph mining for log data. Strata + Hadoop World, San Jose (February
2015)

Machine learning for machine data. Strata Conference, Santa Clara
(February 2014)

Latent Topic Feedback for Information Retrieval. ACM SIGKDD, San Diego (August 2011)

A Framework for Incorporating General Domain Knowledge into Latent
Dirichlet Allocation using First-Order Logic.  IJCAI, Barcelona (July 2011)

Incorporating domain knowledge into topic modeling via Dirichlet
forest priors. ICML, Montreal (June 2009)

Statistical debugging using latent topic models. ECML, Warsaw (September 2007)

PATENTS AND APPLICATIONS

Clustering of structured log data by key schema
United States Patent (11321158)
Udit Saxena, Reetika Roy, Ryley Higa, David M. Andrzejewski, Bashyam TCA

Clustering of structured log data by key-values
United States Patent (11663066)
Udit Saxena, Reetika Roy, Ryley Higa, David M. Andrzejewski, Bashyam TCA

Cardinality of time series 
United States Patent (11182434)
Christian Friedrich Beedgen, David M. Andrzejewski, Weijia Che

Anomaly detection 
United States Patent (10445311B1)
Kumar Saurabh, David M. Andrzejewski, Yuchen Zhao, Christian Friedrich
Beedgen, Bruno Kurtic

Data enrichment and augmentation
United States Patent (11397726)
Christian Friedrich Beedgen, David M Andrzejewski, Benjamin Everette Newton, Kumar Avijit, Stefan Christoph Zier

Logs to metrics synthesis
United States Patent (11042534) 
Christian Friedrich Beedgen, David M Andrzejewski, Benjamin Everette
Newton, Kumar Avijit, Stefan Christoph Zier

Key name synthesis
United States Patent (11481383)
Christian Friedrich Beedgen, David M. Andrzejewski

Visualization tool for system tracing infrastructure events
United States Patent (8464221)
Alice X. Zheng, Trishul A. Chilimbi, Shuo-Hsien Hsiao, Danyel
A. Fisher, David M. Andrzejewski

System and method of drug identification through radio frequency
identification (RFID)
United States Patent Application (11/465993)
Ronald Makin, Kyle Jansson, Silas Zirn, David Andrzejewski, and Timothy Flink.