# Video Editing

Kino AI launched a rebuilt video editor on Hacker News that expresses editing, motion graphics, timeline dragging, teammate comments and 3D scenes as code built on Remotion, so agents such as Codex and Claude Code can work on the same timeline as human teammates. Previously, a raw agent harness could only describe tiny edits in words or one-off interfaces, could not refer to footage shared across devices, and received teammate feedback as Slack screenshots. In the launch video, more than five agents edited the timeline concurrently, one rotoscoping and one color grading all media through LUT support.

Kino AI launched a rebuilt video editor on Hacker News that expresses editing, motion graphics, timeline dragging, teammate comments and 3D scenes as code built on Remotion, so agents such as Codex and Claude Code can work on the same timeline as human teammates. Previously, a raw agent harness could only describe tiny edits in words or one-off interfaces, could not refer to footage shared across devices, and received teammate feedback as Slack screenshots. In the launch video, more than five agents edited the timeline concurrently, one rotoscoping and one color grading all media through LUT support.

On four trailer-editing task types built from 49 real films and their released trailers, decaying an LLM-judged turn-level process reward to zero within the same GRPO setup gives a 45.4% relative gain over outcome-only GRPO on Fine Cut, while the fixed-weight variant scores below outcome-only GRPO on the three shorter task types and truncates 44.1% of rollouts. Process rewards do encourage visual inspection and reasoning early, but continued optimization drives redundant tool calls without a committed result; token-level on-policy self-distillation proves highly sensitive to the teacher's privileged information.

On four trailer-editing task types built from 49 real films and their released trailers, decaying an LLM-judged turn-level process reward to zero within the same GRPO setup gives a 45.4% relative gain over outcome-only GRPO on Fine Cut, while the fixed-weight variant scores below outcome-only GRPO on the three shorter task types and truncates 44.1% of rollouts. Process rewards do encourage visual inspection and reasoning early, but continued optimization drives redundant tool calls without a committed result; token-level on-policy self-distillation proves highly sensitive to the teacher's privileged information.

The work proposes ThinkV2V, a reasoning-driven video-editing framework that explicitly activates multimodal large language model thinking before visual generation: it uses Qwen3-VL-Thinking-8B to perform chain-of-thought reasoning over the source video and instruction and output a refined prompt, injects the hidden states into a Wan-2.1-5B DiT through a learnable-query connector, pairs this with Progressive Curriculum Training and Inference-Time Thinking Scaling, and builds the ThinkV2V-150K dataset and ThinkV2V-Bench; under two judge models, the 5B-scale model achieves the best overall scores on both complex and standard editing scenarios and surpasses several 10B-scale baselines.

The work proposes ThinkV2V, a reasoning-driven video-editing framework that explicitly activates multimodal large language model thinking before visual generation: it uses Qwen3-VL-Thinking-8B to perform chain-of-thought reasoning over the source video and instruction and output a refined prompt, injects the hidden states into a Wan-2.1-5B DiT through a learnable-query connector, pairs this with Progressive Curriculum Training and Inference-Time Thinking Scaling, and builds the ThinkV2V-150K dataset and ThinkV2V-Bench; under two judge models, the 5B-scale model achieves the best overall scores on both complex and standard editing scenarios and surpasses several 10B-scale baselines.