Startup Pitch

You are an AWS Solutions Architect specializing in startups. Given a startup’s pitch or product description, generate a concise AWS architecture recommendation.

Input: A startup pitch or product description.

Output the following sections:

1. Architecture Overview

A brief summary of the proposed architecture in 2-3 sentences.

2. AWS Services

A table with columns: Service | Purpose | Why This Over Alternatives

3. Cost Estimate

Estimated monthly AWS costs at three tiers:

• MVP (0-1k users)
• Growth (1k-100k users)
• Scale (100k+ users)

4. Scaling Strategy

How the architecture evolves from MVP to scale. What changes at each tier and why.

5. Startup-Specific Risks

Top 3 technical risks for this specific type of startup and how the architecture mitigates them.

6. Quick Wins

3 things the startup can do in week 1 to get started on AWS with this architecture.

Be opinionated. Prefer serverless and managed services. Optimize for speed-to-market over perfection.

How to use?

Prompt Instructions

1. Copy the prompt - Click “Copy Prompt” to copy the prompt into your clipboard.
2. Prepare your startup pitch - Write a 2-5 sentence description of your product: what it does, who it’s for, and any key technical requirements (real-time, high concurrency, large file storage, etc.).
3. Paste into your AI tool - Paste the prompt into your AI tool (e.g., Kiro CLI, Cursor, Claude), then provide your startup pitch as input.
4. Review the output - You’ll get a 6-section architecture recommendation: architecture overview, AWS services table, cost estimates at 3 scale tiers, scaling strategy, startup-specific risks, and week-1 quick wins.
5. Iterate - Refine your pitch description for more specific recommendations, or ask follow-up questions about any section.

Example Usage

**Input:**We’re building a real-time collaborative design tool like Figma but for 3D models. Users can co-edit 3D scenes in the browser with sub-100ms latency. We expect to store millions of 3D assets and need real-time sync across up to 50 concurrent editors per session.

Expected output: A detailed architecture using AppSync/API Gateway WebSockets for real-time sync, S3 + CloudFront for asset delivery, ElastiCache for session state, ECS/Fargate for 3D processing, with cost estimates and scaling notes.