Oct 2023 – Present · 2 yrs 5 mos · Remote

• As an advisor to Intent, I contribute my expertise in AI and advanced technology to enhance an intelligent, conversational shopping experience embedded in e-commerce. Intent’s AI-driven platform enables customers to engage naturally with the search function, offering personalized assistance similar to an in-store experience. Through a commerce knowledge graph and continuous adaptation based on product catalogs, reviews, and user feedback, Intent maximizes product discovery, boosts conversion rates, and enhances customer loyalty.

Jul 2022 – Jan 2025 · 2 yrs 6 mos · Seattle, Washington, United States

• Development and applying entertainment graph for multiple Netflix applications including - analytics, slate planning, recommendations etc.

May 2020 – May 2022 · 2 yrs · Seattle, Washington, United States

• As a Tech Lead Manager with extensive experience in AI and ML, I’ve led initiatives at Facebook focused on enhancing user safety and engagement within search, feed, and Marketplace recommendations. In Facebook Search and Feed, I developed models to understand user intent and assess content appropriateness, helping ensure search suggestions and feed recommendations align safely with user queries. Additionally, I improved Marketplace recommendations by connecting users’ short-, medium-, and long-term interests to product graphs, delivering more relevant and personalized results. This blend of research, development, and practical application allows me to transform complex AI models into impactful, user-centered experiences.

Jan 2019 – May 2020 · 1 yr 4 mos

• As a Data Science Manager, I led initiatives to scale and enhance Knowledge Graph integration with a focus on schema mapping, data normalization, and quality assurance. My team implemented semi-automatic schema mapping from web graphs to internal taxonomies and automated normalization for small text, aligning data seamlessly with our knowledge taxonomy. We also developed probabilistic data quality models, ensuring only high-quality data entered the Knowledge Graph. These efforts resulted in a 6X reduction in data onboarding latency, a 4.2X boost in developer efficiency, a 1.3X expansion of the Knowledge Graph, and significant cost savings, driving meaningful improvements in data quality, scale, and operational efficiency.

Jun 2017 – Jan 2019 · 1 yr 7 mos

• 1. Entity Matching Models for Music, Movies, Real-estates, Health, Products etc.
• 2. Supervised, Unsupervised, Semi-supervised and Rule based framework for entity matching.
• 3. Entity Matching data-pipelines on big data processing systems.
• Refer to page (77-100) of tutorial https://kdd2018tutorialt39.azurewebsites.net/KDD%20Tutorial%20T39.pdf

Dec 2015 – Jun 2017 · 1 yr 6 mos

• 1. Entity Matching models for Movies, Music, Real-Estate segment, etc.
• 2. Recall calculation algorithm/infrastructure for entity matching.
• 3. Data-pipelines for entity matching, metrics.

Jun 2014 – Feb 2017 · 2 yrs 8 mos

• 1. Generic ranker for serving Knowledge Graph data.
• 2. Entity search ranker for products(phone, tablets, laptop, and desktop) recommendation.
• 3. Entity search ranker for hotel recommendation.
• 4. Prototyped LinkedIn card suggestion for outlook.
• 5. Search and Lookup API for serving Knowledge Graph data.
• 6. Data-pipelines to publish the data to online store for serving.

Nov 2013 – Jun 2014 · 7 mos

• Microsoft Bing - Indexserve Platform Team. Worked on Ranker cache Manager and Porting the machine learning features.

Jun 2012 – Oct 2013 · 1 yr 4 mos

• First developer of Ranker Shipping Platform for Bing Rankers. My contribution includes designing back-end infrastructure and finally coding E2E with co-workers.
• My contribution also includes working with Redmond Team and developed new data cloud services in the biggest Bing Ranking Infrastructure platform (Aether).

May 2010 – Jun 2012 · 2 yrs 1 mo