We're building the smartest video editing agent that runs autonomously. Under the hood, 25+ AI agents collaborate to deliver perfectly edited videos through prompts. The system can understand existing assets and extract context, generate new pixels/audio with 100+ genAI models, and/or assemble the final composition intelligently and without being restricted to templates. This means more than 80% of the repetitive video edits done automatically.
What You'll Do
* Architect and build multi-agent systems in TypeScript where a primary agent interprets user requests and delegates to specialized sub-agents for tasks like scene composition, visual generation, timeline manipulation, and effect application.
* Design the interface layer between agents and the video editor, enabling agents to read project state, issue commands, and observe results in real time.
* Build sub-agent systems for distinct capabilities such as generating new visual content, applying transformations, adjusting timing, and compositing layers.
* Develop orchestration patterns that allow agents to plan multi-step edits, execute them across the timeline, and recover gracefully when results don't match intent.
* Create memory and context systems that help agents understand the current project state, user preferences, and editing history to make informed decisions.
* Implement feedback loops where agents can preview their changes, evaluate outcomes, and refine their approach before committing edits.
What We're Looking For
Required experience:
* 5+ years of software engineering experience with strong system design skills
* Deep proficiency in TypeScript, including advanced type systems and building robust, maintainable codebases
* Experience with React for building editing interfaces and agent control surfaces
* Hands‑on experience with TypeScript‑based agentic frameworks such as LangChain.js, LangGraph.js, Vercel AI SDK, or custom agent implementations
* Understanding of agent orchestration patterns including hierarchical delegation, plan‑and‑execute loops, and multi‑agent collaboration
* Experience designing APls or command layers that allow programmatic control of complex applications
* Strong prompt engineering skills with experience using function calling and structured outputs for reliable agent behavior
* Comfort working at the intersection of creative tools and Al systems
Nice to have:
* Experience with video editing software, motion graphics, or creative tooling
* Familiarity with generative Al for images or video
* Experience building plugin systems or extensible application architectures
Why Join Us
You’ll work on a video editor built on top of an RLM infrastructure, evaluated as #1 by the fastest growing genAI company in the world. We are experienced founders with a proven track record.
If you’re excited about bringing agentic Al into creative tools, we’d love to talk.
#J-18808-Ljbffr