June 2026 Summaries

Cutsio, founded by Rish Agarwal, is a video search and delivery platform designed for production studios and filmmakers, leveraging Mux Video, Mux Robots, and Mux Player to transform raw footage into searchable, shareable content. Initially an audio-only tool, Cutsio evolved into a comprehensive video platform that addresses key challenges such as organizing scattered video files, making footage easily searchable, and providing an elegant sharing experience. Mux's capabilities in handling large raw files and facilitating seamless video processing enabled Cutsio to develop efficient workflows, including AI-driven tagging and moderation features. By integrating with Mixpeek, Cutsio allows users to search videos using both natural language and visual content recognition, enhancing the discovery of specific moments within extensive footage. The platform also emphasizes professional presentation through its Collections feature, offering a private, theater-style viewing experience that is customizable and mobile-friendly, thus addressing common challenges faced by filmmakers and studios.

Shots is a new feature from Mux Video that creates a manifest of shot boundaries and representative images for video content, forming the basis of the Mux Robots workflow, Find Scenes. The algorithm behind Shots is designed to detect shot boundaries efficiently using minimal resources, prioritizing false positives over false negatives. It operates by analyzing pixel changes between frames, utilizing metrics such as the sum of absolute differences and the mean average frame difference (MAFD) from FFmpeg. Initial approaches involved setting score thresholds and searching for peaks, each with their unique limitations, leading to false positives. The final solution combines both approaches, detecting shot changes when both methods agree, using the last three values for a resource-efficient, one-pass process. While effective, the algorithm struggles with smooth transitions, which can result in false negatives, and with videos containing flashing lights, which can lead to false positives.

Find Scenes is a new workflow added to Mux Robots, designed to transform videos into structured data by identifying and mapping out scene boundaries. This process provides a detailed breakdown of video content, including timestamped ranges, titles, descriptions, and narratives combining audible and visual elements, enhancing the understanding of video by applications. The workflow relies on Shots, a Mux Video primitive that detects visual changes, offering a structured foundation for the analysis. This structured breakdown allows for improved functionalities such as search, clip discovery, navigation, and content browsing, making videos more accessible and functional for AI applications. As an experimental feature, Find Scenes' parameters and pricing are subject to change, and it is priced separately from Mux Robots' usage. Users can access this feature via the Mux dashboard or API, with the option to tailor the workflow through various parameters for a more customized output.

A new video player for React Native apps has been developed, integrating Mux's Swift and Android players with a React Native wrapper that features enhanced UI, Mux Robots, and Mux Data capabilities, all packaged into a single component. This player, installable via npm, leverages the existing Mux native players while adding a React Native wrapper to create a seamless developer experience with support for analytics and digital rights management (DRM). The player architecture consists of three layers: a TypeScript player state, a React view wrapper, and native views, ensuring the React Native API remains simple while maintaining native playback performance. This development aims to refresh the philosophy of existing players like expo-video by incorporating advanced features such as Mux Data metrics and Mux Robots UI, allowing for video summarization, chapter generation, and key moments identification directly from the player. The player supports additional functionalities like signed playback, bandwidth control, clipping, and custom domains, providing a comprehensive tool for React Native developers seeking a robust video player solution with enhanced analytics and user interaction capabilities.

Mux has introduced new features to enhance API rate limit management, providing greater control and reliability for users. Traditionally, Mux applied a single rate limit bucket for all environments, which sometimes led to issues like production disruptions due to overuse in development. To address this, Mux now offers per-environment rate limits, isolating dev, staging, and production environments to prevent one from affecting the others. Additionally, Mux has implemented high and low priority tokens within each environment, allowing users to prioritize crucial API calls, such as video uploads, over less critical tasks like background jobs. The introduction of standard rate limit headers enables users to monitor their API usage in real-time, preventing unexpected request refusals. These updates are designed to optimize API usage and protect end-user experiences, and they are automatically applied to existing environments without requiring user intervention.

Mux Robots has introduced Directives, an orchestration layer designed to simplify the automation and scalability of AI workflows for video assets. Directives allow users to configure and attach specific workflows to video assets, automating tasks like summarizing, translating captions, and generating chapters without requiring additional infrastructure or complex coding. Users can define dependencies for each workflow, ensuring necessary resources are available or generated automatically, and manage how these dependencies are handled if they're missing. Directives can be applied to new assets via the Mux dashboard or API, with future developments planned to enhance their functionality, such as environment-level auto-attach and support for live streams. Mux encourages users to provide feedback on the new feature, which remains free to use in technical preview up to a certain limit, as they work on improving the Directives schema.