EdgeElevate

AI-powered Competitive Intelligence, Orchestrated for Distribution for startups to identify narrative gaps and amplify market presence using LangGraph and Peec AI.

EdgeElevate is a sophisticated intelligence platform that analyzes your competitive landscape, identifies strategic positioning gaps, and generates data-driven content to help early-stage brands win distribution against larger competitors.

1. Pipeline Execution Flow

The system operates as a directed acyclic graph (DAG) powered by LangGraph, transitioning through 4 primary logical stages and 13 specialized nodes.

Frontend Stage	Backend Nodes Involved	Primary Responsibility
Stage 1: Identifying Brand	resolve_project	Mapping user input to Peec AI project entities.
Stage 2: Gathering Intel	fetch_brand_intelligence, fetch_source_intelligence, fetch_actions	Parallel ingestion of raw Peec AI metrics and recommended actions.
Stage 3: Analyzing Gaps	compute_displacement_scores, analyze_narrative, map_source_gaps, analyze_chats	Synthesis of raw data into proprietary competitive displacement scores.
Stage 4: Generating Content	generate_content_opportunities, generate_positioning, generate_linkedin_posts, generate_video_script, assemble_report	Multi-format content generation and final executive report assembly.

System Orchestration Flow

---
config:
  flowchart:
    curve: linear
---
graph TD;
	__start__([START]):::first
	resolve_project(resolve_project)
	fetch_brand_intelligence(fetch_brand_intelligence)
	fetch_source_intelligence(fetch_source_intelligence)
	fetch_actions(fetch_actions)
	analyze_chats(analyze_chats)
	compute_displacement_scores(compute_displacement_scores)
	analyze_narrative(analyze_narrative)
	map_source_gaps(map_source_gaps)
	generate_content_opportunities(generate_content_opportunities)
	generate_positioning(generate_positioning)
	generate_linkedin_posts(generate_linkedin_posts)
	generate_video_script(generate_video_script)
	assemble_report(assemble_report)
	__end__([END]):::last
	__start__ --> resolve_project;
	analyze_chats --> compute_displacement_scores;
	analyze_narrative --> map_source_gaps;
	compute_displacement_scores --> analyze_narrative;
	fetch_actions --> analyze_chats;
	fetch_brand_intelligence --> analyze_chats;
	fetch_source_intelligence --> analyze_chats;
	generate_content_opportunities --> generate_positioning;
	generate_linkedin_posts --> generate_video_script;
	generate_positioning --> generate_linkedin_posts;
	generate_video_script --> assemble_report;
	map_source_gaps --> generate_content_opportunities;
	resolve_project -.-> fetch_actions;
	resolve_project -.-> fetch_brand_intelligence;
	resolve_project -.-> fetch_source_intelligence;
	assemble_report --> __end__;
	classDef default fill:#1e2020,stroke:#adc6ff,color:#e2e2e2
	classDef first fill:#121414,stroke:#4ade80,color:#4ade80
	classDef last fill:#121414,stroke:#ffb4ab,color:#ffb4ab

Loading

Tech Stack: React 19, TypeScript, Tailwind • Python 3.11, FastAPI, LangGraph • Gemini 2.5 Flash • Peec AI • @nivo • Langfuse

2. Detailed Node Logic

Data Ingestion Nodes

• resolve_project: Fuzzy matches user input against Peec AI project list.
• fetch_brand_intelligence: Normalizes Peec AI metrics (Visibility, Sentiment, Position, SOV).
• fetch_source_intelligence: Identifies high-authority domains (Wikipedia, Reddit, YouTube) AI engines cite most.
• fetch_actions: Retrieves recommended action groups (Owned, Editorial, UGC).

Analytical Synthesis Nodes

• analyze_chats: Samples actual AI responses to see exactly how a brand is described compared to competitors.
• compute_displacement_scores: Calculates the proprietary Competitive Displacement Score (CDS).
• analyze_narrative: Extracts framing patterns and identifies "missing narratives."
• map_source_gaps: Categorizes domains into Missing High Authority, Battlegrounds, and Untapped Channels.

Content Generation & Assembly Nodes

• generate_content_opportunities: Ranks 10 content ideas based on displacement potential and effort.
• generate_positioning: Crafts a "Winning Narrative" optimized for AI discoverability.
• generate_linkedin_posts: Drafts 3 posts (Data Insight, Founder Narrative, Product-Led).
• generate_video_script: Creates a script for YouTube, targeting competitor vulnerabilities.
• assemble_report: Synthesizes all analysis into a final executive-ready markdown report.

3. Proprietary Metric: Competitive Displacement Score (CDS)

EdgeElevate quantifies the Ease of Displacement using a weighted formula:

• Visibility Gap (35%): Room for growth vs competitors.
• Sentiment Delta (20%): Quality advantage ("The Quality Wedge").
• Position Proximity (20%): Closeness in AI engine rankings.
• Source Overlap (25%): Shared battleground domains.

Getting Started

Prerequisites

• Node.js (v18+)
• Python 3.11+
• API Keys: OPENROUTER_API_KEY, PEEC_API_KEY

Installation

1. Frontend Setup:

   npm install

2. Backend Setup:

   cd backend
   python -m venv venv
   source venv/bin/activate
   pip install -r requirements.txt

3. Environment: Create a .env in the root and /backend directories with your API keys.

Run Locally

1. Start the Backend:

   cd backend
   uvicorn api_server:app --reload --port 8000

2. Start the Frontend:

   npm run dev

Project Structure

backend/
├── api_server.py           # FastAPI entry point
├── elevate_edge_graph.py   # LangGraph DAG definition
├── models.py               # Pydantic models for structured output
└── requirements.txt        # Python dependencies

src/
├── components/
│   ├── AnalysisFlow.tsx    # Pipeline progress UI
│   ├── Dashboard.tsx       # Main analytics view
│   └── charts/             # @nivo visualization components
└── services/
    └── edgeElevateApi.ts   # SSE streaming client

License

MIT