Project RedSnipe

Analyze Reddit posts to uncover valuable business insights and engagement opportunities using AI-powered analysis.

Comment

We didn't plan to build this.

We were in the middle of a casual convo when our CSO said:

"What if we made an app that could actually handle this for us?"

That one question turned into a full MVP, built in a single 6 hour session using Bolt.new

We didn't brainstorm anything like we usually would. We didn't even whiteboard the idea. We opened Bolt and started building on the spot.

What Inspired Us

This project came straight from a real pain point inside our agency. We weren't trying to build a hackathon entry at that point, we were trying to fix a bottleneck in our sales process. One that, if solved, would give us more time to devote towards building for the Hackathon. That urgency gave us clarity.

What We Learned

  • The best MVPs don't wait, they get used while being built.
  • Real world pain is better than theoretical brainstorming.
  • Recording your build session is worth it. (We filmed everything.)

How We Built It

  • Fully mobile optimized MVP
  • Built for speed and utility, and it also made it look really good!
  • Automated backend logic we could immediately integrate
  • Used live inside our agency before we even "launched"

Challenges

  • Knowing when to stop (we kept wanting to build out features with crazy depth, but knowing when to stop is key)
  • Balancing speed and documentation (filming slowed us slightly)
  • Designing a product that works as intended, AND looks visually appealing and leaves a great first impression upon the user.

Didn't start this to submit it. Now it's our first official hackathon entry. Let's see where this goes.

Built With

Share this project:

Updates

posted an update —

✅ Features & Improvements Added Today

Subreddit Search History Logging

  • Created a Supabase table and SQL logic to track all subreddit searches with user, timestamp, and method.
  • Implemented automatic logging for all subreddit searches (manual, suggestion, etc.).

AI-Powered Search Analytics Dashboard

Built a full-featured analytics page with:

  • Key metrics:
    • Total searches
    • Unique subreddits
    • Preferred search method
    • Active days
  • Interactive charts:
    • Bar, line, and pie charts showing search patterns
  • AI-generated insights and recommendations
  • AI-suggested subreddits based on user history
  • Recent search history display

Navigation & Layout Enhancements

  • Moved the analytics link from the navbar to a prominent button on the Analyzer page (visible only to signed-in users).
  • Made the analytics page full-width, responsive, and visually integrated with the app’s dark theme.
  • Ensured the navbar and user email appear in white on the analytics page for visual consistency.

Chart & UI Polish

  • Fixed label cut-off issues across bar, line, and pie charts.
  • Improved label formatting; added ellipsis for long subreddit names.
  • Updated pie chart to show percentages only, with a color-coded legend displayed below.

Suggested Subreddit Experience

  • Clicking an AI-suggested subreddit on the analytics page now:
    • Navigates to the Analyzer page
    • Auto-fills the input
    • Auto-fetches posts from the selected subreddit

Remixed Response Copy

  • Added a “Copy” button to remixed responses for quick clipboard copying.

Comment Success Predictor

  • Added a “Predict Success” button to each suggested comment.
  • Runs comments through an AI model to predict:
    • Likelihood of upvotes
    • Responses
    • OP attention
  • Displays a color-coded summary and scores.
  • Enhanced AI prompt to include:
    • Post title
    • Post content
    • Subreddit for deeper context
  • Switched to a 1–5 scoring system (5 = strongest).
  • Increased temperature setting for more varied, realistic predictions.

Log in or sign up for Devpost to join the conversation.

posted an update —

Added two GROUND BREAKING Features today:

Subreddit Analysis and User Profile/Post Analysis.

Subreddit Analysis This feature provides a comprehensive overview of a subreddit's potential for business engagement. When you fetch posts from a subreddit or perform a search, you now have the option to "Analyze Subreddit Potential".

How it works: The system takes the fetched Reddit posts and sends them to an AI model (OpenAI). The AI analyzes the collective content of these posts to determine: Overall Sentiment: The general emotional tone of discussions within that subreddit. Key Themes: The most common topics and subjects being discussed. Common Pain Points: Recurring challenges or problems users in that community express. Business Opportunities: Specific areas where your business's services could be relevant, along with their relevance score and matching services. Top Matching Services: Which of your configured services are most aligned with the subreddit's content. Recommended Approach: A strategic suggestion on how to engage with that community. Where to find it: After fetching posts in the "Browse Subreddit Posts" or "Search Reddit Posts" section, a new "Analyze Subreddit Potential" button appears. Once analyzed, the results are displayed above the list of posts and are also saved in your search history for future reference. User Profile/Post Analysis This feature delves into the individual user's online persona to help you tailor your engagement strategy. When you analyze a specific Reddit post, the system now attempts to gather insights about the author of that post.

How it works: If the author of a Reddit post is known (not 'unknown'), the system will: Fetch User Posts: It retrieves the user's last 5 submitted posts from Reddit. Analyze User Sentiment: These posts are then sent to an AI model (OpenAI) for analysis. The AI assesses the user's: Overall Sentiment: Their general emotional tone across their recent posts. Key Themes & Interests: Topics they frequently discuss or show interest in. Communication Style: How they typically interact online. Business Awareness: Their level of understanding or engagement with business-related concepts. Recommended Approach: A tailored suggestion on how to best communicate with this specific user, along with a rationale. Where to find it: After analyzing a Reddit post, the "User Profile Analysis" card will appear within the "Post Analysis" section, providing insights into the post's author. These features are designed to give you a deeper understanding of both the communities and the individuals you might engage with on Reddit, allowing for more targeted and effective outreach.

Log in or sign up for Devpost to join the conversation.