yc0n1c - Startup Advisor - Entrepreneur in Residence

Inspiration

The countless hours COOs and startup executives spend manually gathering customer feedback from scattered sources - App Store reviews, Reddit discussions, support tickets, and more. We realized that an AI-powered Entrepreneur in Residence could automate this entire process, transforming raw feedback into strategic insights.

What it does

Our AI-powered Entrepreneur in Residence continuously monitors 10+ customer feedback sources (App Store, Reddit, Google Play, Twitter, Glassdoor, etc.), analyzes sentiment, identifies trends, and delivers actionable insights through a beautiful real-time dashboard. It combines web scraping (Bright Data), context management (Senso.ai), and data visualization to save executives 5+ hours weekly.

How we built it

• Bright Data: Web scraping infrastructure for collecting real customer feedback
  • Senso.ai Context OS: Storing and managing scraped data with rich metadata
  • Next.js + React: Beautiful, responsive dashboard with real-time updates
  • Express API: Backend server connecting all components
  • Hybrid Approach: 50% real data (App Store, Reddit) + mock data for demo completeness

Challenges we ran into

For web scraping with Bright Data we ran into 2 issues. Out of the box pre-built scrappers didn't cover enough sites for us, and web unlock was too complicated to fully implement for everything during this hackathon. Our solution was to use a hybrid approach, of mostly out of the box solutions, with a couple of custom ones built for sources that were not pre-built ready.

Another major issue was Orkes not working with local data, combined with not being able to get an API key for Senso.ai. Our solution was to store senso.ai data locally, then upload to mongoDB, then have Orkes listen to changes on mongoDB.

Accomplishments that we're proud of

Creating a working dashboard with two distinct type of input(manual input, and web scraping) . Getting an ambient agent to listen to changes, and respond with a trigger. Attempting to tackle a large real world issue, that can continue to be built upon after today. We integrated a basic Vapi voice AI agent.

What we learned

• The power of hybrid approaches when facing API limitations
  • How to build resilient systems with smart fallbacks
  • The importance of context management for AI applications
  • That startups desperately need unified customer intelligence tools

What's next for yc0n1c - Startup Advisor - Entrepreneur in Residence

• Add Mixpanel for product analytics
• Connect to a more sophisticated chatbot that can consume the data and answer questions as an EIR in real time
• Launch as a SaaS product for startups worldwide

Built With

• brightdata
• mixpanel
• mongodb
• orkes
• senso.ai
• typescript
• vapi