Kino AI founder Luke Igel launches Kino video tool for collaboration with Claude Code and Codex
Kino AI founder Luke Igel announced on X on October 29, 2026 that he was launching Kino, saying it supports end-to-end video production with Claude Code and Codex. Users can import materials, obtain highlight clips, edit, and send links for feedback while letting agents participate in the collaboration.
The company’s official website describes Kino as a video editing tool for teams and AI agents to use together, bringing material storage, editing, and review into one place. Editing decisions exist as code on a real timeline, with support for manual adjustments. The system can transcribe material, provide shot-by-shot descriptions, and support natural-language search; it supports local scrubbing of large files and offers templates and parallel editing by multiple agents. Review takes place through links, and comments can be fed back into the timeline.
Igel said that the company previously started with an MIT documentary project, later used the tool on shows such as Survivor and on A24’s Beef, and integrated it with traditional editing software. In 2025, the company rebuilt Kino as a unified editor, betting on editing in code. The official website offers Mac, Windows, and Ubuntu desktop downloads and supports integration with agents such as Claude Code.
The above are public statements by the company and its founder, and the effectiveness of the functions and the scope of actual production use are subject to user testing.
Original sources:
Title: Introducing Kino: make videos end-to-end with Claude Code and Codex. Publisher: Luke Igel (@lukeigel) Publication date: 2026-10-29 URL: https://x.com/lukeigel/status/2115876931749200320 Source nature: original post Verification status: This reading covered the main post and related replies in the thread, verification date: 2026-10-29 Supporting content: product introduction, end-to-end production, importing materials/highlights/editing/feedback/agent collaboration
Title: Kino — AI video editor for teams and agents Publisher: Kino AI Publication date: valid at the time of website access (2026-10-29) URL: https://kino.ai Source nature: official website Verification status: This reading covered the page’s main content, verification date: 2026-10-29 Supporting content: integrated storage/editing/review, agent and team collaboration, transcription and search, timeline and manual adjustment, download platforms
Title: Kino AI: Kino is an AI media manager and video editor for humans and agents. Publisher: Y Combinator Publication date: valid at the time of company page access URL: https://www.ycombinator.com/companies/kino-ai Source nature: accelerator company page Verification status: This reading covered the company description and founder information, verification date: 2026-10-29 Supporting content: YC S23, founder background (MIT, JPL, SpaceX), early product positioning and use cases
ABAB AI Insight
Luke Igel’s background includes computer science training at MIT and engineering experience at NASA JPL and SpaceX. The company began with documentary production experience. The early tool focused on local metadata inference and natural-language retrieval of material to reduce assistant editors’ manual logging work. After the 2025 rebuild, editing decisions were coded so that agents such as Claude Code could operate directly on the timeline while still allowing human manual override.
This path shifts video production from the closed timelines of traditional non-linear editing software to a shared environment that is readable and writable in code and allows multiple agents to work in parallel. Material is still imported by users, while agents handle highlight extraction, editing suggestions, and some effects; humans retain final control. Unlike purely generative video tools, Kino is based on existing material for organization and editing.
The company has previously been deployed in some film and television production and integrated with traditional editors. It is now opening desktop versions to a broader user base and emphasizing compatibility with existing AI coding tools. The result is that agents have been added as collaborators in the editing workflow, while the core decisions remain traceable to editable code and the timeline.
The key points to watch are the proportion of editing actually completed by agents, the degree of interoperability with traditional software, and the retrieval and review efficiency in large media libraries. Current public information is mainly from the founder and the official website.
ABAB News · Cognitive Laws
After editing decisions are coded, AI agents can operate in parallel with humans on the same timeline while preserving manual override capability.
The evolution from a local material indexing tool to a unified editing environment expands agents from retrieval assistance to direct participation in editing.
If a video production tool supports collaboration between humans and agents at the same time, its boundaries depend on code editability rather than full automation.