Summary
Overview
Work History
Education
Skills
Publications
Awards
Patents
Timeline
Generic

Ajay Mysore

San Jose,California

Summary

Data Engineer with deep expertise in distributed systems, SQL processing, and large-scale data infrastructure. Proven track record as a founding member and lead developer, building performant systems from the ground up. Skilled in storage formats, pipeline optimization, and reducing execution engine complexity via SQL interfaces.

Overview

18
18
years of professional experience

Work History

Principle Software Engineer

Workday
Pleasanton, California
04.2015 - Current
  • Enhanced the performance of low-latency SQL queries by 20X on production workloads.
  • Led the migration to Apache Parquet, improving ecosystem compatibility while preserving performance improvements.
  • Developed Spark optimizer rules for SQL window functions, resulting in an 8X performance boost.
  • Implemented SQL Window Functions support, expanding the expressiveness of data transformations and analytics.
  • Improved ingestion performance significantly by eliminating redundant work in complex pipelines.
  • Introduced a caching layer for Parquet scans, significantly reducing latency for repeat queries.
  • Redesigned partition elimination logic to streamline job planning and reduce resource waste.
  • Built a 'Zero Downtime' server restart mechanism, ensuring seamless upgrades in production pipelines.
  • Created a custom debugging tool for production environments, reducing the mean time to recovery for failures.
  • Engineered incremental event-series analysis algorithm for scalable time-series data processing.

Software Engineer

Teradata
San Carlos
02.2010 - 03.2015
  • Designed and implemented over 30 machine learning algorithms on our custom MapReduce-like platform.
  • Designed and developed core features for nPath, Aster's patent-pending pattern-matching SQL-MR operator for large-scale event sequences.
  • Built nTree, a hierarchical data processing operator enabling single-pass traversal of massive datasets.
  • Contributed to the SQL-GR framework (Pregel-like graph processing engine) and implemented APSP-based algorithms for graph analytics.
  • Developed predictive modeling and time-series algorithms, including SAX (Symbolic Aggregate approXimation).

Research Assistant

San Francisco State University
09.2007 - 01.2010
  • Published a paper Advanced Data Mining and Applications Conference 2009.
  • Publication(s): A Semi-Supervised Topic Driven Approach for Clustering Textual Answers to survey questions. Hui Yang, Ajay Mysore, and Sharonda Wallace. The fifth international conference on Advanced Data Mining and Application (ADMA 2009).
  • Large-Scale Graph Analytics in Aster 6: Bringing Context to Big Data Discovery. David Simmen, Karl Schnaitter, Jeff Davis, Yingjie He, Sangeet Lohariwala, Ajay Mysore, Vinayak Shenoi, Mingfeng Tan, Yu Xiao. The seventh conference on Very Large DataBases (VLDB 2014).
  • Patent: Pattern recognition across multiple input datasets, IDR number DN13-1027. This patent is filed based on a feature of nPath designed to find patterns across multiple database tables.

Education

Master of Science - Computer Science

San Francisco State University
San Francisco, CA
01.2010

Skills

  • Data processing
  • SQL optimization
  • Machine learning
  • Data architecture
  • Software development
  • Performance optimization
  • Team leadership
  • Event-series analysis
  • Data structures
  • Parquet format optimization
  • Big data analytics
  • Distributed systems
  • SQL Internals

Publications

  • A Semi-Supervised Topic Driven Approach for Clustering Textual Answers to survey questions, Hui Yang, Ajay Mysore, and Sharonda Wallace, The fifth international conference on Advanced Data Mining and Application (ADMA 2009)
  • Large-Scale Graph Analytics in Aster 6: Bringing Context to Big Data Discovery, David Simmen, Karl Schnaitter, Jeff Davis, Yingjie He, Sangeet Lohariwala, Ajay Mysore, Vinayak Shenoi, Mingfeng Tan, Yu Xiao, The seventh conference on Very Large DataBases (VLDB 2014)

Awards

  • Teradata R&D Excellence awards, Distinguished Award, 2014, for nPath Multi-Input and Symbol-Based Lag Features for Aster.
  • Recognized in an Architect-in-Training program, 2024, highlighting leadership potential and system design expertise.

Patents

Pattern recognition across multiple input datasets, DN13-1027, This patent is filed based on a feature of nPath designed to find patterns across multiple database tables.

Timeline

Principle Software Engineer

Workday
04.2015 - Current

Software Engineer

Teradata
02.2010 - 03.2015

Research Assistant

San Francisco State University
09.2007 - 01.2010

Master of Science - Computer Science

San Francisco State University

Similar Profiles

CREATE PROFILE