Justin Talbot Email & Phone Number

San Francisco, Ca, Us

San Francisco, Ca, Us

Seattle, Wa, Us

Seattle, Wa, Us

Seattle, Wa, Us

Member of the Office of the CTO's leadership team. Advise Chief Technical Officer on research and advanced development directions.Help senior leadership develop yearly strategic roadmap. Prepare and present roadmap summaries to Tableau's senior leadership team and to the board of directors.Evaluate M&A opportunities.Perform architectural reviews of major projects across the development organization. Led a cross-cutting team in creating a plan to address persistent performance problems.Developed and prototyped a quarterly planning process to replace the existing annual process.

Seattle, Wa, Us

Led multiple applied R&D efforts focused on developing solutions to some of the key challenges in Tableau's core analytic area. Led teams of 3--12 members in brainstorming, prototype development, and evaluation. Wrote initial problem definitions and resulting roadmaps. Presented results to senior management.Consulting architect for Tableau's query pipeline, coordinating and reviewing designs across 5 feature teams. Responsible for aligning short-term development with long-term vision.Former Technical Lead, Query Pipeline Team. Led an 8 person cross-functional development team that took one of my research prototypes and built it into a complete replacement for Tableau's query backend. Replaced an ad hoc query generator with a general-purpose query language that enabled new features and faster development. Implemented a query rewriter and optimizer to improve performance.Designed and delivered substantial extensions to Tableau's analytic capabilities, including level-of-detail expressions, improved sorting and filtering behavior, and support for modeled data sources.Research PublicationsAnushka Anand and Justin Talbot. Automatic Selection of Partitioning Variables for Small Multiple Displays. InfoVis 2015.Justin Talbot, Vidya Setlur, Anushka Anand. Four Experiments on the Perception of Bar Charts. Infovis 2014.Organized Visualization in Practice workshop at VisWeek 2015 and 2016. InfoVis 2015 and 2016 Organizing Committee member.2 patents (+2 pending)

Stanford, Ca, Us

Developed a high performance version of the R statistical language. Involved research in JIT compilation strategies, virtual machine design, and data structures for scalable, high-performance data analytics. Published as:Justin Talbot, Zachary DeVito, and Pat Hanrahan. Just-in-time Length Specialization of Dynamic Vector Code. ARRAY 2014.Justin Talbot, Zachary DeVito, and Pat Hanrahan. Riposte: A Trace-Driven Compiler and Parallel VM for Vector Code in R. PACT 2012.Available here: https://github.com/jtalbot/riposteResearched algorithms for automatic design of graphical data presentation. Published as:Justin Talbot, John Gerth, and Pat Hanrahan. An Empirical Model of Slope Ratio Comparisons. Infovis 2012.Justin Talbot, John Gerth, and Pat Hanrahan. Arc Length-based Aspect Ratio Selection. InfoVis 2011.Justin Talbot, Sharon Lin, and Pat Hanrahan. An Extension of Wilkinson’s Algorithm for Positioning Tick Labels on Axes. InfoVis 2010.Researched MapReduce performance on multicore and distributed machines and developed Phoenix++, a highly efficient C++ MapReduce library for multicore processing. Published as:Justin Talbot, Richard M. Yoo, and Christos Kozyrakis. Phoenix++: Modular MapReduce for Shared-Memory Systems. Second International Workshop on MapReduce and its Applications (MAPREDUCE) 2011.Available here: https://github.com/kozyraki/phoenixTyson Condie, Neil Conway, Peter Alvaro, Joseph M. Hellerstein, John Gerth, Justin Talbot, Khaled Elmeleegy, Russell Sears. Online Aggregation and Continuous Query support in MapReduce. SIGMOD 2010 Demo.Researched analytics and visualization of RDF data. Published as:Bryan Chan, Leslie Wu, Justin Talbot, Mike Cammarano, Pat Hanrahan. Vispedia: Interactive Visual Exploration of Wikipedia Data via Search-Based Integration. IEEE Information Visualization, 2008.

Growing out of my Ph.D. work, I helped found 386 Gates, Inc; developed business plans, built prototypes, pitched investors and customers, interviewed potential hires. Shortly before incorporation, I pulled the plug due to differences of opinion on company organization and product direction with my co-founder.The company eventually became 0xdata, then H2O.ai. I had no part in that good work other than owning some (likely highly diluted) founder shares.

Redmond, Washington, Us

Developed a visualization tool, EnsembleMatrix, for human-guided improvements to ensemble machine learning algorithms. Planned and executed a user study of the system, demonstrating that, with a human in the loop, our system could achieve substantially higher accuracy classification results than the state of the art automated method. Published as:Justin Talbot, Bongshin Lee, Desney Tan, and Ashish Kapoor. EnsembleMatrix: Interactive Visualization to Support Machine Learning with Multiple Classifiers. CHI 2009.1 patent.

Walldorf, Bw, De

Developed a prototype trend-tracking tool for internal use at SAP. Portions of the project were presented to SAP’s CEO and board members. Project incorporated research directions from information foraging and automatic text summarization in a community-driven website geared towards tracking industry trends. Published as:Omar Alonso and Justin Talbot. Structuring collections with Scatter/Gather extensions. SIGIR 2008: 697-698.

Redmond, Washington, Us

Worked on Visio and OfficeArt, Office’s core rendering library (as part of a “SWATteam” fixing high priority problems late in the development cycle)Gained experience with GDI/GDI+, OLE, remote development, and large-scale commercial software development practices---C++, code review, version control, debugging, and working with program managers and testers

Helped start-up develop the first release of the T-Splines Maya plugin in C++. Researched generalizations of t-splines (a mathematical 3D surface representation) to arbitrary topology meshes.

Provo, Ut, Us

Researched and published statistical techniques for reducing error in renderedimages. Implemented global illumination systems using path tracing, bi-directional path tracing, and Metropolis Light Transport. Published as:Justin F. Talbot, David Cline, and Parris K. Egbert. Importance Resampling for Global Illumination. Eurographics Symposium on Rendering 2005, pages 139-146.David Cline, Justin F. Talbot, and Parris K. Egbert. Energy Redistribution Path Tracing. ACM Transactions on Graphics (SIGGRAPH 2005 Proceedings).David Cline, Parris K. Egbert, Justin F. Talbot, and David L. Cardon. Two Stage Importance Sampling for Direct Lighting. In Rendering Techniques 2006 (Eurographics Symposium on Rendering), pp. 103-113, (June 2006)Researched terrain rendering techniques and optimizations. Led the redesign of an interactive 3D terrain visualization engine for commercial release using OpenGL. Published as:Justin F. Talbot and Parris K. Egbert. VTVS - A Complete Implementation of a Virtual Terrain Visualization System. IASTED VIIP 2003, Benalmadena, Spain, September 2003.

Ph.D., Computer Science

M.S., Computer Science

B.S. Cum Laude, Computer Science