Developer for AI Video Creation App

I'm looking for a developer to help me build an app that automatically creates faceless YouTube videos. It combines two things: AI video generation — creating the video clips themselves with AI, like in this video: https://youtu.be/HPWc58OwRz0 VidRush-style automation (vidrush.ai) — automatic b-roll selection and assembly into a finished, ready-to-upload video IMPORTANT — where the real work is. In order of priority: 1. AI video generation at a reasonable cost with good quality. This is a core pillar of the app: generating the actual clips with AI video models. I'm NOT going to tell you which providers or models to use — that's exactly what I want YOU to tell me. Which ones you'd pick, why, what each scene type needs, how you'd optimize the generation prompts, how you'd avoid wasted generations, and what the real cost per finished video would be. The quality-vs-cost trade-off is the heart of the product, and your answer here is how I'll judge if you know this space. 2. B-roll selection engine (VidRush-style). Automatically picking the right footage for each part of the script: deciding when to use AI-generated video, AI images, or stock footage, and making sure the visual actually matches what the narration is saying at that moment. This is what separates a good tool from a bad one. 3. The assembly/montage engine — audio perfectly synced with the clips. This is CRITICAL. Each clip must start and end exactly where it should relative to the voiceover, with word/phrase-level precision. Cuts on the right beat, no drift over long videos. Tell me how you'd solve the audio-visual alignment and what you've built like this before — this is the #1 skill I'm hiring for. Lower priority (already solved, don't focus your proposal on this): script generation and AI voiceover. That part is easy and takes me no time — I don't need help there. Good to know: I already have working Python scripts that produce videos end-to-end (AI visuals + stock b-roll + audio-synced assembly). You can build on top of them or propose a better architecture — your call, explain your reasoning. Budget: Open — it depends on what you can actually deliver. We'll start with a small paid milestone (MVP) and continue from there if the quality is right. Please include a rough estimate for an MVP in your proposal. In your proposal, please include: Links or samples of videos your code generated AND assembled (AI video generation + editing pipeline, not just clips made manually with an AI tool) — I want to see both generation quality and sync quality Which AI video providers/models you'd use, why, and the approximate cost per finished minute How you'd approach the audio-clip synchronization, in 2-3 sentences Start your proposal with the word "PIPELINE" so I know you read this Long-term collaboration possible if the first milestone goes well.