A problem isn't truly solved until it's solved for all. Googlers build products that help create opportunities for everyone, whether down the street or across the globe. Bring your insight, imagination and a healthy disregard for the impossible. Bring everything that makes you unique. Together, we can build for everyone.\n\nCheck out our career opportunities at goo.gle/3DLEokh

Indian Institute of Technology, Bombay

University of Washington

Seasoned TLM (Tech Lead Manager), leading an engineering team and collaborating across product, design and engineering to deliver on core Assistant user-journeys that have O(100M) MAUs. Also responsible for hiring, coaching and managing a team of SWEs.

Currently leading the effort for integrating Gemini, Google's LLM technology into Google Assistant for the Google Home ecosystem and launching new voices for Google Assistant. This is a x-PA project with Home, Speech etc. teams.

TLM (Tech Lead Manager) for Assistant Productivity journeys portfolio. 

Previously, TL for Assistant Notes and Lists journeys portfolio, including Google Shopping List, 0P Infrastructure, integrations with Google Keep and other partners. Led and worked on several projects:
• A multi-year, x-PA project to consolidate Assistant's previously-fragment offerings. 
• Live data migration of O(100M) lists and notes from multiple backends to Google Keep, signifying the product and tech stack
• Multi-quarter project to integrate Google Keep and other partners with Assistant.

Taught the 'Machine Learning Crash Course' and 'Machine Learning For Product Teams' courses at Google. Also involved with hiring and interviewing.

Received ~50 Peer Bonuses / Spot Bonuses in recognition of going above and beyond.

Software Engineer

SanskritSahitya.org is one of my passion projects that tries to combine my interest in Linguistics and NLP with my interest in languages and Sanskrit in particular. It's a free website aimed at organizing Sanskrit literature and associated information, such as syntactical / morphological analysis for texts, computerized metrical analysis, access to and rich search functionality etc., in a structured and easily navigable form, thus furthering the study and analysis of Sanskrit texts.

All data present on the website is made freely available through https://github.com/sanskritsahitya-com/data

Creator

• As part of the founding "Attribute Extraction" team, used NLP techniques to systematically extract attributes from the product feed sent by BloomReach customers. Filed (and granted) 3 patents in this space
• Developed a suite of web tools for QAing product attributes and enabling quick integration for new customers
• On own initiative, created S3Browser, a web based file browser and viewer for Amazon S3
• Actively involved in hiring—vetting and interviewing engineering candidates and mentoring new hires
• Won 2nd prize in company Hackathon (2015), organized company Hackathon (2016)
• Won peer awards in eight consecutive quarters

Member of Technical Staff

1. Created a model to find hierarchical structure in unsupervised fashion from raw query data received from Amazon’s analytics portal; proposed model achieves significant improvement over current system

2. Developed high precision machine-learning and rule-based models for categorizing books in the Kindle catalog; tested model over 160K books, achieving 10x improvement in tagging and 4x improvement in readership coverage.