Dec 2020 – Jan 2026 · 5 yrs 1 mo · Palo Alto, California, United States

• Building modern AWS Lake house architecture which combines the goodness of Data lakes and Data warehouses along with unified data governance across different query engines, transactional data management and seamless data movement - all in a server-less architecture with scalable compute and storage for capacity and performance needs.
• AWS Glue Data Catalog acts as a central meta-store for all data lake and database style assets including structured, unstructured objects, stored procedures, VIEWs, connections. The Catalog is a large distributed multi-tenant server-less meta store catering to high throughput low latency requests in data path.
• Leads the engineering for Iceberg REST Catalog and federation to external data sources - Catalog federartion, Credential federation and Permissions federation
• AWS Lake Formation helps:
• Build secure data lakes and modern data mesh architecture - Ingest data from varied data sources (S3, relational, no-SQL) into data lake, crawl, catalog and prepare for analytics.
• Centrally define and manage access controls
• Enforce data classification and fine grained access
• Enable continuous data management, time travel, and storage optimization
• Enable federated data lakes with cross-account sharing

Jan 2017 – Dec 2020 · 3 yrs 11 mos · Mountain View

• FlashBlade - Unified File and Object Scaleout Distributed All-Flash Platform. Developed and led NAS storage features for distributed filesystem - FileSystem Snapshots, Replication, NFSv3, v4.x file protocols, Directory Services(LDAP, AD, OUD), Security features (Kerberos, SSL, TLS), Access Control Lists (ACLs - NFS/SMB) for granular permission evaluation, Scale-out Multi-chassis (150 blades) - with focus on performance and capacity scale in highly distributed platform.

Dec 2014 – Dec 2016 · 2 yrs · San Jose

• Delivered software for Flash based Storage System involving core Flash Management. Led projects on Garbage collection, Block Pick Algorithms, Inter process Communication infra, Memory management infra (RDMA, Persistent NVRAM, NVDIMM). Designed Meta-Data Management for Solid State Storage System – Map Journal, Meta Block management.

Jun 2013 – Dec 2014 · 1 yr 6 mos · San Jose

• Designed and Developed software for SAN(Storage Area Network)
• Control and Management Plane for all flash single controller and dual controller storage array
• Distributed Systems Software for High Availability in Dual controller all flash enterprise storage array
• Implemented Inter process communication infra for inter controller communication
• Expertise in Flash Management, Snapshots, Clones, Control Plane management, High Availability features, Distributed Systems Software Development

Jun 2012 – Jun 2013 · 1 yr · San Jose

• Responsible for design, development and debugging of user space MAC drivers for Nexus 7000 series switches
• Implemented 40G/100G breakout feature. (100G/40G physical port can be broken out into various 10G/40G combinations)
• Designed and Implemented drop log infrastructure for user space drivers
• Worked on Software design for one of the next generation system-on-chip ASIC (MAC / Port-Logic Chiplet).

May 2011 – Aug 2011 · 3 mos

• Involved in Systems Management feature design and metrics driven decisions
• Understood the technology drive and customer need for the feature by doing market research analysis on the product
• Designed and implemented providers for remote systems management
• Worked with winrm client and powershell cmdlets to talk to openwsman
• Worked with openwsman and sfcb opensource implementations
• Developed MS powershell scripts for security and test applications

Aug 2010 – Dec 2011 · 1 yr 4 mos

• Developing Algorithms in C++ and Python to sequence the genetic data and handling intensive data sets using different Algorithms and Data Structures
• Analyze and calculate the heterogeneity among the genes through clustering algorithms
• Implemented an efficient Cache Replacement technique – Shepherd cache in C language using the Simple Scalar Simulator and compared the results with traditional LRU strategy and calculated the performance improvement. Developed Machine Learning Algorithms in C++ / Python to sequence the genetic data and built statistical models around the date for inference (Neural Network, Decision Tree, Nearest neighbor, Bayesian classifiers, clustering).
• Skill Set
• Programming in C++, python on Linux environment
• Data Structures, Algorithms, Machine Learning Algorithms

Nov 2008 – Apr 2009 · 5 mos · Chennai, India

• Completed project on Wireless Sensor Networks – “Energy Estimation of energy efficient MAC Protocol for Wireless Sensor Networks”
• Implemented the S-MAC (Sensor-MAC) protocol- An energy efficient protocol, to come up with an energy estimation formula for the WSN using tinyOS and NesC
• Skill Set:
• C, Computer Networks & Protocols, Operating Systems, Wireless Sensor node architecture