ShrijaAI

srijaAI

Inspiration Content creation today is largely driven by intuition rather than data. Creators often struggle to identify what will trend, when to post, and how to maximize engagement. While platforms provide analytics, they are reactive—not predictive.

We were inspired to solve this gap by building a system that doesn't just analyze past performance but predicts future virality. Additionally, growing concerns around data ownership and platform dependency pushed us to design a decentralized AI solution where creators retain control over their insights and identity.

What it does SrijaAI acts as an AI-powered co-creator and strategist for content creators and influencers. It: Analyzes real-time global and niche-specific trends Predicts content performance using a Virality Factor Score (VFS) Recommends what to post in the next 48 hours Generates content assets like hooks, scripts, captions, thumbnails, and CTAs Personalizes suggestions based on audience behavior and creator niche In short, it transforms content creation from guesswork into a data-driven, predictive workflow.

How we built it We designed SrijaAI as a modular, decentralized AI system: Trend Analysis Engine: Scrapes and processes Web2 + Web3 social signals (hashtags, engagement spikes, topic velocity) AI Models: Used NLP and predictive modeling to calculate the Virality Factor Score (VFS) Recommendation Engine: Combines trend data + user profile to generate personalized content strategies Content Generator: AI modules for generating scripts, hooks, and creatives Decentralization Layer: Built on ICP to ensure data ownership, transparency, and censorship resistance Frontend: Clean, creator-friendly dashboard for actionable insights Challenges we ran into Accurate Virality Prediction: Building a reliable scoring system required balancing multiple dynamic factors like timing, audience behavior, and trend lifecycle Real-time Data Processing: Handling fast-moving trend data efficiently without latency Personalization: Making recommendations truly niche-specific rather than generic Decentralized Architecture: Integrating AI workflows with decentralized infrastructure (ICP) without compromising speed Noise vs Signal: Filtering meaningful trends from hype and spam Accomplishments that we're proud of Successfully built a working Virality Factor Score (VFS) model Created a 48-hour predictive content engine Integrated AI-generated content creation with trend analysis Designed a decentralized creator-first architecture Delivered an end-to-end system that goes from trend detection → prediction → content generation What we learned Predicting trends is not just about data volume but data relevance and timing Creators value actionable insights over raw analytics Combining AI + decentralization unlocks new ownership models Personalization is the key differentiator in creator tools Simplicity in UX is critical—even for complex AI systems

What's next for SrijaAI Improve VFS accuracy using advanced ML models and feedback loops Expand multi-platform integrations (Instagram, YouTube, X, Web3 platforms) Launch creator analytics dashboard with deeper insights Introduce collaborative AI agents for teams and agencies Build a marketplace for creator strategies and AI models Scale the decentralized infrastructure for global adoption