How we brought multimedia search to Dropbox Dash
May 29, 2025Jay Howe

Summary
This post details the engineering behind adding multimedia search (images, video, audio) to Dropbox Dash. It covers the challenges of storage and compute costs, relevance, and responsiveness for media files. The solution involves indexing lightweight metadata (file path, title, EXIF) using the Riviera framework, backfilling the index with raw content for preview generation, and optimizing storage. At query time, it matches input against metadata, applies custom logic for geographic references, and tokenizes camel case/hyphenated strings. Previews are generated just-in-time on the read path to balance latency and resource efficiency.
More from the Universal Content Processing Platform trail
- Evolving our calendar assistant Reclaim to be AI-native without starting over
- How our universal content processing platform Riviera evolved for AI and beyond
- How we used DSPy to turn AI evaluations into better responses in Dash chat
- How we optimized Dash's relevance judge with DSPy
- Using LLMs to amplify human labeling and improve Dash search relevance