RivalFree: Multi-Agent Competitive Analysis Engine Powered by Gemma & SerpApi

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass

What I Built

RivalFree is an autonomous, AI-driven competitive intelligence and feature strategy platform designed to get founders, developers, and product managers off their screens and back into the real world.

Normally, performing competitive research means sitting in front of a laptop for days—manually scouring competitor landing pages, parsing pricing tiers, and reading endless forum threads to spot feature gaps. RivalFree fixes this screen fatigue.

By submitting a simple product concept, RivalFree spins up a parallel multi-agent pipeline that scans live web data via SerpApi, maps competitor threats, surfaces customer pain points, and generates a first-to-market execution roadmap in seconds. It converts 40 hours of screen-grinding research into a 15-second report so founders can close their laptops and touch grass.

Demo

• Live Platform: https://rivalfree.sumeet.app/
• Video Demo: https://drive.google.com/file/d/1O7F8qUIeTIyeG6rhMfmHRjNuGditmt9E/view?usp=sharing

Code

sumeet57 / RivalFree

How I Built It

RivalFree is engineered with a modern, reactive multi-agent stack:

• Agent Framework & Open AI Integration: Built on @langchain/google-genai and flexible LangChain orchestrators designed to integrate seamlessly with open-weight models (such as Gemma, Llama 3, and Qwen via HuggingFace or Ollama local inference).
• Live Search & Grounding: Integrated SerpApi directly into the query-planning agent to pull live organic SERP results and public forum complaints, eliminating static model hallucinations.
• Real-Time Streaming Engine: Uses Node.js, Express, and Socket.IO to stream agent reasoning steps ("Query Planning", "SERP Scanning", "Parallel Synthesis") directly to a reactive terminal UI.
• Persistent Data & Session Management: MongoDB / Mongoose for storing projects, feature void matrices, and persistent chat sessions.
• Deployment Infrastructure: Containerized using Docker Compose and deployed on AWS EC2 behind an Nginx reverse proxy with SSL certificate management via Certbot.

Why Does Open Innovation Matter?

Competitive analysis should not be locked behind expensive enterprise paywalls or proprietary, closed-box algorithms.

Open innovation and open-weight models ensure that solo founders and independent developers can audit the agent pipeline, run market intelligence locally without leaking sensitive product ideas to closed third-party APIs, and adapt open-weight reasoning models to their specific niche. Open-source AI democratizes strategic intelligence for everyone.

Prize Categories

• Main Track: Week 1 - Touch Grass
• Featured Category: Best Use of Gemma (Running open-weight reasoning pipelines)
• Partner Category: Best Use of SerpApi (Grounding multi-agent market research in live Google SERP data)
• Partner Category: Best Use of MongoDB Atlas (Using Atlas as the primary data layer for storing user projects, feature void matrices, and persistent chat sessions)