Automating Content Creation with an AI-Powered Platform
A deep dive into the full-stack SaaS platform built to accelerate content operations for creators and marketers, streamlining everything from video repurposing to multi-platform ad campaigns.
Project Impact
The platform successfully delivered an automated content pipeline, a reliable scheduling system, and an intuitive ads "wizard," leading to significant improvements in operational efficiency.
From Challenge to Solution
The Challenge
- Creators were bogged down by repetitive tasks: clipping videos, designing thumbnails, and scheduling posts across multiple platforms.
- Existing tools were siloed, creating a disjointed and inefficient workflow between editors, schedulers, and ad dashboards.
- Teams lacked a unified dashboard to track performance across both organic content and paid advertising campaigns.
Our Solution
- Built an end-to-end platform using AI for content analysis, clip generation, and thumbnail creation, automating the most tedious tasks.
- Integrated key platforms (YouTube, TikTok, Google Ads) into a single dashboard for streamlined content management and distribution.
- Developed a simplified "wizard" for Google Ads, leveraging AI to suggest assets and dramatically reduce campaign setup time.
Advanced Technology Stack
System Design & Architecture
High-Level Architecture
Entity-Relationship Diagram
Core Platform Features
AI Content Repurposing
Users paste a YouTube URL to initiate an AI analysis that automatically suggests engaging short-form clips. The integrated editor allows for fine-tuning start/end points and selecting platform-specific aspect ratios before processing.
Unified Scheduling & Library
A central library provides a consolidated view of all generated clips with robust filtering and search. The scheduler allows users to plan and queue posts across multiple platforms, visualizing the content calendar at a glance.
AI Thumbnail Studio
The studio generates high-quality thumbnails from a simple AI prompt and an optional reference image, offering customizable templates to ensure brand consistency and save valuable design time.
Technical Challenges & Solutions
Video Repurposing Latency
Solution: Implemented an asynchronous job queue, parallelized clip generation, and enabled incremental delivery. A fallback FFmpeg process was also developed for critical performance needs.
API Quotas & Timeouts
Solution: Engineered fine-tuned upload timeout configurations and wrapped all API calls with an adaptive backoff algorithm to dynamically manage request rates and prevent exhaustion.
Concurrent Cron Job Safety
Solution: A transactional database locking mechanism was implemented. A worker job atomically "claims" a task, preventing other workers from processing the same item simultaneously.
Dual Google Ads Auth Models
Solution: Designed a dual-mode architecture to handle token acquisition dynamically based on whether agency-level (MCC) or individual user (BYO) credentials were used.
Duplicate Campaign Names
Solution: Before creation, the system performs a duplicate check via GAQL. If a collision is detected, it automatically appends a unique suffix and retries the request.
Multiple TikTok Upload Modes
Solution: The upload module was built to support both chunked and pull-from-URL protocols, with intelligent logic for backoff polling, task cancellation, and resuming failed uploads.
Frequently Asked Questions
What was the primary goal of the TopFloor Trends project?
The main goal was to create a unified SaaS platform to drastically reduce the time and complexity creators and marketers face. We aimed to automate content repurposing, streamline multi-platform scheduling, and simplify ad campaign creation using AI.
How does the AI-powered clip generation work?
A user provides a YouTube video URL. The system's AI model analyzes the content to identify the most engaging segments. It then returns structured data with suggested clips, topics, and timestamps, which are presented to the user for final selection and refinement.
How were technical challenges like API limits handled?
We engineered the system with robust error handling. This included fine-tuned upload timeout configurations and an adaptive backoff algorithm that dynamically manages request rates. This approach prevents API quota exhaustion and ensures system stability, especially during high-load operations.
How was the system deployed and maintained?
The application is deployed on Vercel. We utilized Vercel's scheduled serverless functions (Cron Jobs) to run every 15 minutes for processing the content queue. These functions were configured with a max duration of 300 seconds to reliably handle long-running tasks like video uploads.
