Venky Manicks

Background

About Venky Manicks

I lead the team of engineers responsible for developing multimodal search ranking models/algorithms that power Google Lens and Multimodal GenAI experiences. My team is responsible for:* Multimodal Search Retrieval and Ranking: For multimodal (image-based) queries, we retrieve/rank images and web documents from multiple 100B+ corpora using both traditional ML algorithms and deep models.* RAG and Rich Content in GenAI: For multimodal (image-based) queries, our ranking stack provides RAG capabilities for Google's AI Mode and LLM-orchestrated ranking of rich (non-text) content within AI responses. Our work directly impacts 1.5B+ MAU, and enables flagship experiences including:* Multisearch (https://blog.google/products/search/multisearch/): Search using images and text at the same time * Lens AI Overviews (https://blog.google/products/google-lens/lens-on-ios-ai-overviews/) Previously, I led the ranking team responsible for improving visual and content diversity in Google's Image Search.