Rahul Potharaju Email & Phone Number

San Francisco, Ca, Us

San Francisco, Ca, Us

Redmond, Washington, Us

Engineering leader for Azure Synapse Analytics working on next-generation experiences for Apache Spark managing a talented, diverse and geo-distributed team of 40 people. Responsible for reliability, experience, supportability and price/perf charters of the Apache Spark engine in Azure Synapse Analytics.Responsible for the account of the largest customer of Spark in Azure Synapse Analytics. This includes consulting with them on their architectures, unblocking them as needed and helping with capacity planning and product escalations.Responsible for growth hacking prototypes and establishing many cross-company/university collaborations including with NVIDIA, Databricks, Imperial College of London, and University of Illinois at Urbana Champaign.

Redmond, Washington, Us

Founding member of Azure Synapse Analytics and Engineering leader for Apache Spark in Azure Synapse Analytics. I am responsible for getting the Microsoft-powered Spark Runtime (w/ performance enhancements, Hyperspace Indexing, Delta Lake and a plethora of other data connectors) in the hands of thousands of our customers across all of Azure, in a reliable manner and making sure it stays reliable and trustworthy. In addition, I am responsible for making the user experience around using Spark the best-in-class so users can focus more on their business and less on worrying about how to debug Spark.Architect and Lead for the GPU-hardware accelerated Spark Runtime for Azure Synapse Analytics. Founding member of .NET for Spark (https://github.com/dotnet/spark) and Hyperspace Indexing Sub-system for Spark (https://github.com/microsoft/hyperspace - also check out our VLDB'21 paper). I have given several Data+AI Summit talks on these - please feel free to visit their catalog or YouTube to view these videos.

Redmond, Washington, Us

Founding member of Helios, the hyperscale indexing system for cloud and edge. Helios is a distributed, highly-scalable system used at Microsoft for flexible ingestion, indexing, and aggregation of large streams of real-time data that is designed to plug into relational engines. The system collects close to a quadrillion events indexing approximately 16 trillion search keys per day from hundreds of thousands of machines across tens of data centers around the world. Helios use cases within Microsoft include debugging/diagnostics in both public and government clouds, workload characterization, cluster health monitoring, deriving business insights and performing impact analysis of incidents in other large-scale systems such as Azure Data Lake and Cosmos. Helios also serves as a reference blueprintfor other large-scale systems within Microsoft. Checkout our VLDB'20 paper here: http://www.vldb.org/pvldb/vol13/p3231-potharaju.pdfAcolyer's Helios: hyperscale indexing for the cloud & edge – part 1 https://blog.acolyer.org/2020/10/26/helios-part-1/Acolyer's Helios: hyperscale indexing for the cloud & edge – part 2 https://blog.acolyer.org/2020/11/02/helios-part-ii/

Redmond, Washington, Us

Redmond, Washington, Us

West Lafayette, In, Us

West Lafayette, In, Us

Work on modeling and building a framework consisting of an extension module to ns3 simulator that allows worm simulations. In addition, work on adaptive access control in smartphones, leveraging behavior patterns for classifying entities into various threat levels in a P2P system and social network privacy related aspects.

West Lafayette, In, Us

Work on modeling and simulating malware propagation through wireless networks using techniques from Mathematical epidemiology.

Redmond, Washington, Us

(Latest first)- Work on evaluating new datacenter architectures, statistical modeling and prediction of failures in network devices, traffic modeling for anomaly detection in real-time, and building large scale real-time datacenter network monitoring systems. - Work on applying techniques from natural language processing to analyze network trouble tickets and software bug reports for extracting trends.- Work on building a search interface to enable search of configuration information present within natural text in expert written articles (Microsoft Knowledge base articles).- Work on building a real-time event monitoring system that helps troubleshoot datacenter related outages. This involved designing the system architecture, developing the event extraction pipeline, API design and visualization. - Co-Architect and Lead Developer of NetSieve (http://netsieve.info) - Work on building a real-time event monitoring system that helps troubleshoot datacenter related outages. This involved designing the system architecture, developing the event extraction pipeline, API design and visualization. In addition, work on investigating the utility of natural language processing on software bug repositories.- Work on building network diagnostic tools and algorithms to help troubleshoot datacenter problems at Microsoft Research, Redmond. In addition, work on developing techniques for identifying and mitigating availability issues in today's data center networks is being carried out. - Work on designing and building a reliable control plane from the ground-up for futuristic networks at Microsoft Research, Redmond. Specifically, working on developing new techniques for identifying and mitigating availability issues in today's data center networks was carried out.

Chicago, Illinois, Us

Designed and implemented the Companion Device Framework inside Motorola Applied Research (2010) which was being used by four different teams consisting of twenty full-time researchers and developers. The framework enabled end-user data collection for use in various analytics and recommendation systems inside Motorola.

Chicago, Illinois, Us

- Work on designing, simulating and prototyping new neighborhood caching mechanisms for next generation television set-top boxes. The system was built using Java and implements simulation models based on various network growth models and real-world models based on a real dataset.- Design and deployment of multi-dimensional k-means clustering using Google's MapReduce and Apache Hadoop.- Analytics Framework development and deployment on Google's Android Platform. The application provides programmers with an API to communicate with the underlying OS and enables data collection tasks for research purposes. An Android application, leveraging this framework's capabilities, that evaluates the effects of ontological aspects of data presentation on users was developed and deployed to a set of users at Rice University.

Evanston, Il, Us

Work on a distributed system monitoring system for Planet-Lab test-bed using PHP + MySQL. In addition, initiated various large scale Internet measurement tasks.

Evanston, Il, Us

Evanston, Il, Us

Teaching Assistant for the course "Introduction to C++ for Non-Programmers"

This was a non-profit startup (2004-2008) that I founded which was similar to StackOverflow (http://stackoverflow.com) in spirit. As part of this job title, I successfully managed a team of 75 team members and accumulated 10K+ community members in a span of three years. During its lifespan, the company has won two national level accolades for being one of the largest student projects.

Evanston, Illinois, Us

Developed a customized parser for a particular business case using PERL

Chicago, Illinois, Us

Work on designing and implementing a graphic visualizer for Motorola's P2K Operating System.

Phd, Computer Science

Ms, Computer Science

Btech, Electronics And Communication Engineering