Powered by the Northstar engine developed at MIT and Brown University

Einblick is based on 6 years of research done at Massachusetts Institute of Technology (MIT) and Brown University as part of the DARPA Data Driven Discovery of Models (D3M) program and several NSF grants.




Image

Key Features


The most natural interface:
The first human-centric interface to analytics.

The fastest engine:

The first data science tool based on a progressive computation engine.
The smartest assistants:
The first interactive Auto-ML and data mining tool.
Cloud-native:
Easy deployment in your public or private cloud environment.



The most natural interface

The first Visual Data Computing interface

The first Visual Data Computing interface that enables playful interaction with data and supports desktops, large interactive whiteboards, and tablets and features real-time collaborative editing.


Watch the video demo

Image



Visual Data Computing combines the best aspects of workflow engines, BI tools, programming environments, and collaborative tools.


Image
Data Workflow Engine
Image
Business Intelligence Software
Image
Data Programming Interface
Image

Optimization
What-if


Image



Progressive Computation Engine


The first progressive computation engine


All interactions take under a second regardless of the data size and results are then continuously refined in the background.





Philipp Eichmann, Emanuel Zgraggen, Carsten Binnig, Tim Kraska: IDEBench: A Benchmark for Interactive Data Exploration. SIGMOD Conference 2020: 1555 1569
“Delays of 500ms incurred significant costs, decreasing user activity and data set coverage while reducing rates of observation, generalization and hypothesis.”

Zhicheng Liu, Jeffrey Heer
The Effects of Interactive Latency on Exploratory Visual Analysis. IEEE Trans. Vis. Comput. Graph. 20(12): 2122 2131 2014



Image

The smartest assistants


The first interactive data mining and Auto-ML tool

Our award-winning smart assistants help to automatically find statistically significant insights and build models in sub-seconds rather than hours. It enables everyone to do tasks only data scientists can do now.





In 80% of the cases, Einblick’s models were better than DARPA’s expert solution.

Zeyuan Shang, Emanuel Zgraggen, Benedetto Buratti, Ferdinand Kossmann, Philipp Eichmann, Yeounoh Chung, Carsten Binnig, Eli Upfal, Tim Kraska: Democratizing Data Science through Interactive Curation of ML Pipelines. SIGMOD Conference 2019: 1171-1188



Image

Cloud-enabled



Einblick can be deployed in the public or private cloud. We also allow airgap installation for the most critical applications.

Adding data can be as easy as dropping a file or you can connect easily connect to your favorite data sources with a click of a button. No need for expensive loading and data preparation steps. Einblick's progressive sampling engine takes care of the rest.




Based on years of cutting-edge research


Einblick is powered by the Northstar engine, which was originally developed by MIT and Brown University over 6 years of research. The results have been published in various top-tier systems, machine-learning, and human-computer interactions conferences and journals



Zeyuan Shang, Emanuel Zgraggen, Benedetto Buratti, Ferdinand Kossmann, Philipp Eichmann, Yeounoh Chung, Carsten Binnig, Eli Upfal, Tim Kraska
Democratizing Data Science through Interactive Curation of ML Pipelines
SIGMOD2019
Tim Kraska
Northstar: An Interactive Data Science System
PVLDB 11(12): 2150-2164 (2018)
Yeounoh Chung, Sacha Servan-Schreiber, Emanuel Zgraggen, Tim Kraska
Towards Quantifying Uncertainty in Data Analysis & Exploration
IEEE Data Eng. Bull.41(3): 15-27 (2018)
Carsten Binnig, Benedetto Buratti, Yeounoh Chung, Cyrus Cousins, Tim Kraska, Zeyuan Shang, Eli Upfal, Robert Zeleznik, Emanuel Zgraggen
Towards Interactive Curation & Automatic Tuning of ML Pipelines
DEEM@SIGMOD 2018: 1:1-1:4
Yeounoh Chung, Michael Lind Mortensen, Carsten Binnig, Tim Kraska
Estimating the Impact of Unknown Unknowns on Aggregate Query Results.
ACM Trans. Database Syst. 43(1): 3:1-3:37 (2018)
Emanuel Zgraggen, Zheguang Zhao, Robert C. Zeleznik, Tim Kraska
Investigating the Effect of the Multiple Comparisons Problem in Visual Analysis
CHI2018: 479
Alex Galakatos, Andrew Crotty, Emanuel Zgraggen, Carsten Binnig, Tim Kraska
Revisiting Reuse for Approximate Query Processing
PVLDB 10(10): 1142-1153 (2017)
Emanuel Zgraggen, Alex Galakatos, Andrew Crotty, Jean-Daniel Fekete, Tim Kraska
How Progressive Visualizations Affect Exploratory Analysis
IEEE Trans. Vis. Comput. Graph. 23(8): 1977-1987 (2017)
Carsten Binnig, Lorenzo De Stefani, Tim Kraska, Eli Upfal, Emanuel Zgraggen, Zheguang Zhao
Toward Sustainable Insights, or Why Polygamy is Bad for You
CIDR 2017
Yue Guo, Carsten Binnig, Tim Kraska
What you see is not what you get!: Detecting Simpson's Paradoxes during Data Exploration
HILDA@SIGMOD 2017: 2:1-2:5
Tim Kraska
Approximate Query Processing for Interactive Data Science
SIGMOD Conference 2017: 525
Zheguang Zhao, Lorenzo De Stefani, Emanuel Zgraggen, Carsten Binnig, Eli Upfal, Tim Kraska
Controlling False Discoveries During Interactive Data Exploration
SIGMOD Conference 2017: 527-540
Philipp Eichmann, Emanuel Zgraggen, Zheguang Zhao, Carsten Binnig, Tim Kraska
Towards a Benchmark for Interactive Data Exploration
IEEE Data Eng. Bull. 39(4): 50-61 (2016)
Muhammad El-Hindi, Zheguang Zhao, Carsten Binnig, Tim Kraska
VisTrees: fast indexes for interactive data exploration
HILDA@SIGMOD2016: 5
Andrew Crotty, Alex Galakatos, Emanuel Zgraggen, Carsten Binnig, Tim Kraska
The case for interactive data exploration accelerators (IDEAs)
HILDA@SIGMOD 2016: 11
Andrew Crotty, Alex Galakatos, Emanuel Zgraggen, Carsten Binnig, Tim Kraska
Vizdom: Interactive Analytics through Pen and Touch
PVLDB 8(12): 2024-2027 (2015)
Evan R. Sparks, Ameet Talwalkar, Daniel Haas, Michael J. Franklin, Michael I. Jordan, Tim Kraska
Automating model search for large scale machine learning
SoCC 2015: 368-380
Emanuel Zgraggen, Robert C. Zeleznik, Steven M. Drucker
PanoramicData: Data Analysis through Pen & Touch
IEEE Trans. Vis. Comput. Graph. 20(12): 2112-2121 (2014)
Tim Kraska et al
MLbase: A Distributed Machine-learning System
CIDR 2013