Inspiration
Having a good idea is only the beginning. Turning it into engaging content means researching examples, understanding an audience, writing a script, planning shots, and editing everything together. For independent creators and small businesses, that can become a second job.
We built Yap — Good ideas, made REEL. to bring those steps into one creative workspace. Our goal is to help people turn their ideas, photos, and product knowledge into original content while staying in control of the result.
What it does
Yap connects inspiration, research, and video creation.
Creators can start with an idea or their own photos, explore public examples, and choose a creative direction. Yap helps develop that direction into an editable script and shot plan, followed by reviewed video production and a downloadable reel.
The research workspace combines what people post with how accounts and posts connect. An interactive graph lets creators explore topics, estimated communities, and observed interactions, inspect supporting sources, and save creative decisions to inform their scripts.
Guided and Director views support different working styles: one helps creators develop an idea through conversation, while the other exposes the script, shots, references, and production controls.
How we built it
We built the frontend with React, TypeScript, and Vite, and the backend with Node.js and Express. Shared Zod schemas validate information exchanged between the applications.
Our research orchestration uses LangChain Deep Agents. A coordinator plans the work and delegates to specialists for source research, network analysis, and creative strategy. Shared research files carry each specialist’s findings into the next step.
The analysis pipeline combines text similarity with observed network relationships. Model integrations support interpretation and creative proposals, while media tools including Sharp and FFmpeg support image preparation and video assembly.
Creator review, version checks, and explicit production approval help keep the final output aligned with the current creative plan.
How Yap relates to the tracks
Entertainment and Media
Yap centers on storytelling and original media creation. It helps creators translate inspiration into concrete choices about hooks, narrative structure, visuals, and pacing, then develop those choices into a reel.
Arlequin AI — Social Media Analysis: Combining Content & Connectivity Patterns
Yap analyzes both message content and connections between actors. Text similarity and observed interactions jointly inform estimated communities and research insights. The graph makes those relationships explorable, while source references let users inspect the evidence behind them.
LangChain — Best use of deepagents
Yap uses Deep Agents for a coordinated research workflow involving planning, specialist delegation, tools, and shared artifacts. Source collection, network analysis, and creative strategy form dependent steps that produce an attributed creative brief.
Amazon — Best Practical AI Innovation
Yap applies AI to a recurring creative task: turning raw material into usable content. Its workflow connects research, planning, revision, and production around a clear target user. Editable proposals and creator feedback make the assistance responsive to the project’s context.
Peri — Forward Deployed Challenge
Yap’s practical focus aligns with building around real business workflows. A local shop or creator can bring their own material and communication goals into a tool designed to make content production easier to adopt.
Challenges we faced
Connecting research to creation required careful boundaries. External examples can inspire a format or hook, but their claims should not become facts about someone else’s product.
We also had to keep scripts, scenes, approvals, and renders consistent as creators revised their work. Coordinating agents introduced another challenge: each specialist needed clear responsibilities, bounded tool access, and dependable handoffs.
What we learned
We learned that useful AI workflows depend on clear intermediate results, visible evidence, and opportunities for human judgment. We also learned that combining content and connectivity provides richer context for creative research.
What’s next
We want to refine Yap through feedback from creators and businesses, expand brand customization, improve production quality, and help users connect creative decisions with the results of their published content.
Built With
- claude
- langchain
- openai
- xai
