Manoj Pawar

2025 - Present: AI Search @ Meta SuperIntelligence Labs Developing a highly personalized RAG/AI search system designed to create personal SuperIntelligence by leveraging all of Meta’s content. 2021-2025: Recommendations @ Instagram Spearheaded the creation of industry-leading personalized recommendation systems for Instagram Reels and Growth. Under my leadership, the team drove approximately 25% overall growth for Instagram by strategically leveraging Reels and other growth initiatives - successfully engaging non-power users and fostering long-term retention on the platform. 2015 - 2021: Personalized Search Ranking @ Facebook Led multiple initiatives to enhance Facebook Blue App search ranking by leveraging advanced personalization signals. Successfully improved key metrics, including increasing search click-through rate by over +xx%, search quality by +yy%

Areas: Visual Search Platform, Relevance, Information Retrieval, Distributed Computing, Lucene, Hadoop

Co-creator of Learning To Rank (LETOR) System for Telenav Local Search. Implemented features which are responsible for CTR improvement of 6% in Keyword queries and 3% in Categorical queries. Worked on Lucene, Log Mining, Analytics using Hadoop & Cascading. Creator of "Did you mean" feature for Telenav local search. Implemented language model (NLP) along with various distance based algorithms to correct misspelled queries.

Worked on mSpot movies streaming server. Worked on mSpot music Genius feature for recommendation of similar songs using various music features.

Built key areas of the Viewzi.com search platform, including work on the View Picker, Memcaching technique and Executive Information System (EIS). Designed and implemented various ways to retrieve data from different data providers. Analyzed end-user needs and software requirements to determine feasibility of view design on viewzi platform. Internship: Designed and developed a Viewpicker algorithm for ranking of views on context of given search.

Worked on a project involving enhancements of several Ecommerce based ‘Private Label Credit Card’ applications such as e-Apply, e-Security, e-Pay. This involved following the waterfall model of the SDLC, covering requirement analysis, designing and coding. Acquired expertise in JAVA, Struts, Shell-Scripting and Database design.

Thesis: “Visual Querying and Content Based Searching for 3D Human Motions” Implemented visual query interface for drawing motion queries and designed a content based search algorithm for efficient retrieval of 3D Human Motions. Publications: Manoj Pawar, Gaurav N. Pradhan, Kang Zhang, Balakrishnan Prabhakaran, “Content Based Querying and Searching for 3D Human Motions”, International Proceedings of ACM Multimedia Modeling Conference (MMM) 2008, pp. 446-455, Kyoto, Japan, January 9-11,2008.

Applied Learning (Data Mining), Information Retrieval, Natural Language Processing Research Paper: Automated Summarization of Restaurant Reviews http://nlp.stanford.edu/courses/cs224n/2010/reports/mpawar-dmallya.doc