Skip to content

omertt27/Founder-Copilot

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

21 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

πŸš€ Founder Copilot

**AI-Powered Startup Helper ### πŸ€– Amazon Nova Models Used

Model Bedrock Inference Profile ID Generation Role Speed
Nova 2 Lite us.amazon.nova-2-lite-v1:0 Gen 2 ⚑ Default β€” all 5 features ~4-8s
Nova Pro us.amazon.nova-pro-v1:0 Gen 1 ✨ High-quality generation option ~6-10s
Nova Premier us.amazon.nova-premier-v1:0 Gen 1 πŸ† Most powerful (selectable) ~8-12s
Nova Micro us.amazon.nova-micro-v1:0 Gen 1 πŸš€ Intent classification in Auto-Detect agent ~2-4s

Turn your startup idea into a comprehensive plan, technical architecture, development backlog, investor-ready pitch deck, and go-to-market strategy β€” all powered by Amazon Nova AI through Amazon Bedrock.

#AmazonNova | Category: Agentic AI | Hackathon: Amazon Nova AI Hackathon on Devpost

Amazon Nova AI Nova 2 Python React


🎬 Demo Video

πŸŽ₯ Watch the Demo on YouTube β€” ~3 minutes | #AmazonNova

Demo flow: Launch app β†’ Startup Plan generation β†’ Tech Architecture β†’ GitHub Issues β†’ Pitch Deck β†’ Marketing Strategy β†’ Demo Mode (no AWS) β†’ Streaming SSE in real time


🎯 Try It Instantly (No AWS Needed!)

Founder Copilot includes a Demo Mode that works without any AWS credentials. Just clone, install, and run:

git clone https://github.com/omertt27/Founder-Copilot.git
cd Founder-Copilot/backend && pip install -r requirements.txt && python run.py &
cd ../frontend && npm install && npm run dev

πŸ§ͺ Demo Mode activates automatically when no AWS credentials are configured. It returns realistic sample outputs so you can explore every feature immediately.


✨ Features

Feature Description Avg. Time
πŸ’‘ Startup Plan Comprehensive strategy with product description, target users, MVP features, 6-month roadmap, and success metrics ~6-10s
πŸ—οΈ Tech Architecture Full tech stack recommendation with database schema, API endpoints, security considerations ~7-12s
πŸ“‹ GitHub Issues 15-20 prioritized development issues with estimates, acceptance criteria, and labels ~8-14s
🎀 Pitch Deck 12-slide investor-ready pitch deck outline with talking points and visual suggestions ~8-12s
πŸ“£ Marketing Strategy Full go-to-market plan with target audience analysis, channel breakdown, content strategy, and budget allocation (CMO-in-a-box) ~6-10s
πŸ€– Auto-Detect Smart agent that detects your intent and routes to the right feature (uses Nova Micro for fast classification) ~10-15s
πŸ“‘ Streaming Real-time Server-Sent Events for progressive output rendering Instant chunks

οΏ½οΏ½ Amazon Nova Models Used

Model Bedrock Inference Profile ID Role Speed
Nova 2 Lite us.amazon.nova-2-lite-v1:0 ⚑ Default β€” Gen 2 fast reasoning, great quality ~4-8s
Nova Pro us.amazon.nova-pro-v1:0 ✨ Gen 1 high quality, comprehensive outputs ~6-10s
Nova Premier us.amazon.nova-premier-v1:0 πŸ† Most powerful (selectable) ~8-12s
Nova Micro us.amazon.nova-micro-v1:0 πŸš€ Intent detection in Auto-Detect agent ~2-4s

πŸ“Έ Screenshots

Feature Selection & Input

Input Panel

Generated Output β€” Startup Plan

Startup Plan Startup Plan continued Startup Plan continued

Generated Output β€” GitHub Issues

GitHub Issues GitHub Issues continued GitHub Issues continued

πŸ“· To add screenshots: Run the app locally (npm run dev) and take screenshots at 1280Γ—800 or wider. Save them to docs/screenshots/ using the filenames above. See docs/SCREENSHOTS.md for the full guide.


πŸ“ Example Outputs (One-Liners)

Here's what each feature generates from a simple prompt like "AI meeting assistant that records, summarizes, and creates action items":

Feature Sample Output Highlights
πŸ’‘ Startup Plan Product description, 3 user personas, 7 MVP features, 6-month roadmap with milestones, 7 KPIs, 3 risk mitigations
πŸ—οΈ Tech Architecture Next.js + FastAPI + PostgreSQL stack, system architecture diagram, 4-table database schema, 10 API endpoints, security & scaling notes
πŸ“‹ GitHub Issues 17 prioritized issues (#1 Project Setup β†’ #17 Rate Limiting), each with priority, estimate (S/M/L), acceptance criteria, and labels
🎀 Pitch Deck 12 slides (Cover β†’ The Ask), $37B problem stat, TAM/SAM/SOM market sizing, competitive matrix, $2.5M ask with fund allocation
πŸ“£ Marketing Strategy 6-section GTM plan, channel matrix (SEO/Paid/Social/Email), content calendar, persona targeting, 90-day launch roadmap, KPIs

πŸ“Š Performance Metrics

Metric Value
Startup Plan generation ~4-8 seconds (Nova 2 Lite default)
Tech Architecture generation ~4-8 seconds (Nova 2 Lite default)
GitHub Issues generation ~5-10 seconds (Nova 2 Lite default)
Pitch Deck generation ~4-8 seconds (Nova 2 Lite default)
Marketing Strategy generation ~4-8 seconds (Nova 2 Lite default)
Intent detection (Auto) ~1-2 seconds (Nova Micro)
Demo mode response ~1-3 seconds (simulated)
Average tokens per response 1,800 - 3,000 tokens
Rate limit 10 requests/minute per IP
Total codebase 32+ files, 3,100+ lines

πŸ› οΈ Tech Stack

Backend

  • Python 3.11+ with FastAPI
  • Amazon Bedrock (Nova 2 Lite, Nova Pro, Nova Premier, Nova Micro)
  • boto3 for AWS integration
  • slowapi for rate limiting
  • Pydantic v2 for request/response validation

Frontend

  • React 18 with Vite 6
  • react-markdown for rendering
  • Framer Motion for animations
  • react-icons (Heroicons v2) for UI icons

οΏ½οΏ½ Quick Start

Prerequisites

  • Python 3.11+
  • Node.js 18+
  • (Optional) AWS Account with Bedrock access for real Nova AI responses

1. Clone the Repository

git clone https://github.com/omertt27/Founder-Copilot.git
cd Founder-Copilot

2. Backend Setup

cd backend
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt

3. Configure Environment

cp .env.example .env

Option A: With AWS credentials (full power):

AWS_ACCESS_KEY_ID=your_access_key_here
AWS_SECRET_ACCESS_KEY=your_secret_key_here
AWS_REGION=us-east-1

Option B: Demo mode (no credentials needed):

DEMO_MODE=true
# Or just leave AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY empty

4. Start Backend

cd backend
python run.py

Backend runs at: http://localhost:8000 API docs at: http://localhost:8000/docs

5. Frontend Setup

cd frontend
npm install
npm run dev

Frontend runs at: http://localhost:5173


πŸ“ Project Structure

Founder-Copilot/
β”œβ”€β”€ backend/
β”‚   β”œβ”€β”€ app/
β”‚   β”‚   β”œβ”€β”€ __init__.py
β”‚   β”‚   β”œβ”€β”€ main.py              # FastAPI app entry point
β”‚   β”‚   β”œβ”€β”€ config.py            # Environment & settings (+ demo mode)
β”‚   β”‚   β”œβ”€β”€ nova_client.py       # Amazon Nova/Bedrock client
β”‚   β”‚   β”œβ”€β”€ prompts.py           # All prompt templates
β”‚   β”‚   β”œβ”€β”€ models.py            # Pydantic schemas
β”‚   β”‚   β”œβ”€β”€ routes.py            # API endpoints (7 routes)
β”‚   β”‚   └── demo_responses.py    # Mock data for demo mode
β”‚   β”œβ”€β”€ run.py                   # Server runner
β”‚   β”œβ”€β”€ requirements.txt
β”‚   β”œβ”€β”€ .env                     # Your credentials (gitignored)
β”‚   └── .env.example             # Template
β”‚
β”œβ”€β”€ frontend/
β”‚   β”œβ”€β”€ src/
β”‚   β”‚   β”œβ”€β”€ components/          # React components (7 files)
β”‚   β”‚   β”œβ”€β”€ services/api.js      # API client with SSE streaming
β”‚   β”‚   β”œβ”€β”€ App.jsx              # Main app (state, routing, history)
β”‚   β”‚   β”œβ”€β”€ App.css
β”‚   β”‚   β”œβ”€β”€ index.css            # Global design system
β”‚   β”‚   └── main.jsx             # Entry point
β”‚   β”œβ”€β”€ index.html
β”‚   β”œβ”€β”€ package.json
β”‚   └── vite.config.js
β”‚
β”œβ”€β”€ docs/
β”‚   β”œβ”€β”€ screenshots/             # Demo screenshots
β”‚   └── SCREENSHOTS.md           # Screenshot recording guide
β”œβ”€β”€ README.md
β”œβ”€β”€ PROJECT_REPORT.md            # Full system architecture report
└── .gitignore

πŸ”Œ API Endpoints

Method Endpoint Description Rate Limited
POST /api/generate/startup-plan Generate startup plan βœ… 10/min
POST /api/generate/tech-architecture Generate tech architecture βœ… 10/min
POST /api/generate/github-issues Generate GitHub issues βœ… 10/min
POST /api/generate/pitch-deck Generate pitch deck βœ… 10/min
POST /api/generate/marketing-strategy Generate marketing strategy (CMO) βœ… 10/min
POST /api/generate/auto Auto-detect & generate βœ… 10/min
POST /api/generate/stream/{feature} Streaming generation (SSE) βœ… 10/min
GET /health Health check (includes demo_mode flag) ❌

Response Format

{
  "feature": "startup_plan",
  "content": "# πŸ’‘ Startup Plan: ...",
  "model_used": "nova2lite",
  "tokens_used": 1847,
  "generation_time": 8.42,
  "demo_mode": false
}

πŸ§ͺ Demo Mode

Founder Copilot includes a built-in demo mode for trying the app without AWS credentials:

Aspect Details
Activation Automatic when AWS_ACCESS_KEY_ID is empty, or set DEMO_MODE=true
Outputs Realistic, pre-written sample content for all 5 features
UI Indicator Yellow "Demo Mode" badge in header + banner on outputs
Latency 1-3 seconds simulated delay for realistic feel
Streaming Simulated chunk-by-chunk streaming

This makes the app instantly demoable for hackathon judges without needing AWS access.


🎬 Demo Examples

Try these prompts to see what Founder Copilot generates:

Example 1: AI Meeting Assistant

πŸ’‘ Startup Plan: "AI-powered meeting assistant that records meetings, 
generates summaries, and creates action items automatically"
β†’ Generates: 6-section plan with 7 MVP features, 6-month roadmap, and KPIs

Example 2: Freelancer SaaS

πŸ—οΈ Tech Architecture: "SaaS tool for freelancers to track time, 
send invoices, and manage clients"
β†’ Generates: Full stack recommendation, database schema, 10 API endpoints

Example 3: AI Documentation Tool

πŸ“‹ GitHub Issues: "Developer tool that uses AI to automatically 
write documentation from code comments"
β†’ Generates: 17 prioritized issues with estimates and acceptance criteria

Example 4: EdTech Platform

🎀 Pitch Deck: "Personalized AI tutor that adapts to each student's 
learning style and pace"
β†’ Generates: 12-slide deck with market sizing, competitive matrix, and financial projections

Example 5: B2B SaaS

πŸ“£ Marketing Strategy: "Project management tool for remote engineering teams"
  Target Audience: "CTOs and engineering managers at 10-100 person startups"
  Budget: "$5,000/month"
β†’ Generates: Full GTM plan with channel breakdown, content calendar, paid vs organic split, and 90-day launch roadmap

πŸ”’ Security

Feature Implementation
Credentials AWS keys in .env (gitignored), never in code
Rate Limiting 10 requests/minute per IP (slowapi)
Input Validation Pydantic v2 schemas with min_length constraints
CORS Configurable allowed origins
Async Safety boto3 calls run in thread pool executor
Error Handling Structured error responses (422, 429, 500)

πŸ‘¨β€πŸ’» Built For

Amazon Nova AI Hackathon β€” Demonstrating the power of Amazon Nova 2 Lite (Gen 2) alongside Nova Pro, Nova Premier, and Nova Micro for real-world startup tooling.

What Makes This Special

  • 🧠 Nova 2 Lite (Gen 2) β€” Default model for all 5 features; Nova Micro for fast intent detection
  • ⚑ Production-ready β€” Rate limiting, async safety, input validation, error handling
  • 🎨 Beautiful UI β€” Dark theme with feature-colored accents, Framer Motion animations, responsive design
  • πŸ§ͺ Instantly demoable β€” Works out of the box with demo mode (no AWS needed)
  • πŸ“Š Metrics built-in β€” Generation time, token count, and model info displayed on every output
  • πŸ“£ Full founding team simulation β€” Startup Plan (CEO), Tech Architecture (CTO), GitHub Issues (Engineering Lead), Pitch Deck (Investor Relations), Marketing Strategy (CMO)

πŸ† Hackathon Submission

Category

Agentic AI β€” Founder Copilot uses Amazon Nova's reasoning capabilities to simulate a full AI founding team. The Auto-Detect agent classifies user intent (via Nova Micro) and routes to the correct specialized Nova 2 generation agent, each with a distinct expert persona and structured multi-section output.

Project Description (copy-paste this into the Devpost submission form)

Founder Copilot is an AI-powered startup helper agent that turns any startup idea into a complete founder package in seconds. Powered by Amazon Nova 2 Lite (Gen 2, default) and Nova Pro via Amazon Bedrock, it simulates a full founding team:

  • πŸ’‘ CEO β†’ Startup Plan (strategy, roadmap, KPIs)
  • πŸ—οΈ CTO β†’ Technical Architecture (stack, schema, API design)
  • πŸ‘©β€πŸ’» Engineering Lead β†’ GitHub Issues (prioritized dev backlog)
  • 🎀 Investor Relations β†’ Pitch Deck (12-slide investor outline)
  • πŸ“£ CMO β†’ Marketing Strategy (GTM plan, channels, 90-day roadmap)

An Auto-Detect agent (Nova Micro) classifies the user's natural language input and routes it to the right specialist automatically. All generation uses real-time Server-Sent Events (SSE) streaming for a live typewriter experience. Includes instant Demo Mode (no AWS required) so judges can test every feature without credentials.

Built with FastAPI + React + Amazon Bedrock. #AmazonNova

How It Uses Amazon Nova

Nova Model Bedrock Inference Profile ID Generation Role in App
Nova 2 Lite us.amazon.nova-2-lite-v1:0 Gen 2 Default generation model β€” all 5 features
Nova Pro us.amazon.nova-pro-v1:0 Gen 1 High-quality generation option
Nova Premier us.amazon.nova-premier-v1:0 Gen 1 Most powerful (selectable)
Nova Micro us.amazon.nova-micro-v1:0 Gen 1 Intent classification in Auto-Detect agent

Testing Instructions for Judges

# Clone the repo
git clone https://github.com/omertt27/Founder-Copilot.git
cd Founder-Copilot

# Start backend (Demo Mode works without AWS credentials)
cd backend && pip install -r requirements.txt && python run.py &

# Start frontend
cd ../frontend && npm install && npm run dev
# β†’ Open http://localhost:5173

πŸ§ͺ No AWS account needed β€” Demo Mode activates automatically if no credentials are set. For real Nova AI responses, add your AWS credentials to backend/.env.

Grant repo access for judging (if repo is private): Share with testing@devpost.com and Amazon-Nova-hackathon@amazon.com


πŸ“ License

MIT License β€” Built with ❀️ and Amazon Nova AI

About

Founder-Copilot is a open source project which aims to help beginner founders to get free services from specialized ai agents

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors