Jul 2021 – May 2023 · 1 yr 10 mos

• Led a team as part of building a B2B e-commerce marketplace under the Ads vertical at Google. Owned the design, development and launch of features such as:
• Product Bookmarks: Save products on the search results page.
• User Workspace: Organize products into custom lists for comparison and project-based purchases. Supports users adding comments to products and lists, and share lists with other users.
• Search Filters: Filter products based on attributes such as delivery date and return policy.
• Also owned the deployment and monitoring aspects of the marketplace serving stack.

Jan 2020 – Jan 2021 · 1 yr

• Led the design and development of data mining service for the product. Functionalities include dynamic injection to identify, rank and re-inject dynamic values from response into subsequent requests (e.g. IDs) to test a service; and comparison rules to identify dynamic and reordered values during original and mock response to render smart comparison diffs.
• The product is open sourced at https://github.com/mesh-dynamics/api-studio.

Jan 2015 – Jan 2020 · 5 yrs · Bengaluru, Karnataka, India

• Part of Advanced Development Centre group: a team with a charter to propose, incubate, validate, and productize novel ideas in emerging technologies. Worked on:
• Big Data Acceleration using In-Storage Compute:
• Implemented a predicate filter server on HGST Open Ethernet Drive that accepted data sources and filter predicates to return pre-filtered data to the client. Enhanced Cloudera Impala’s execution engine to offload filter operations to the drive. By moving computation closer to the data, the enhanced Impala engine achieved significant average speedups of 20% on big data workloads.
• Containerized Rapid Blockchain Deployment Framework:
• Responsible for backend development of a blockchain-as-a-service platform that enabled users to quickly provision a cluster of interconnected Ethereum blockchain nodes using a web UI. Supported parsing of functions in the deployed smart contract and auto-generate corresponding REST end point code. Provisioning was done using containers which were instrumented to export system-level metrics for monitoring.
• Swarm Learning -- Privacy-Preserving Decentralized Machine Learning
• Worked on conceptualization, design and development of the Swarm Learning product for decentralized privacy-preserving ML. Worked with multiple engineers to deliver multiple components including
• the Swarm library (core parameter merge logic), file server (model exchange infra), and ML/blockchain network infra.
• Published the Swarm Learning whitepaper describing the use cases and a high-level architecture of the product

Jan 2012 – Jan 2012 · 0 mo

• Worked on the MAC layer of the embedded networking stack for Long Term Evolution (4G) compatible
• mobile chipsets. Designed and implemented the protocol stack based on 4G RFC standard specifications

Jan 2011 – Jan 2011 · 0 mo

• Came via acquisition of Networks business of Motorola. Was part of Operating Systems and Device Drivers Team.

Jan 2009 – Jan 2011 · 2 yrs

• Contributed to the initial days of development of 4G systems, particularly in the Mobility Management Entity. Work involved concepts from parallel processing, high-availability, socket programming, and load balancing.