Technical Architecture
Shipwright AI is built on a modern, cloud-native stack designed for high availability, security, and developer productivity. Every component is chosen for reliability and maintainability.
System Architecture
Next.js / React / TypeScript / Tailwind CSS
Security Boundary — HTTPS / OAuth / RBAC
OAuth 2.0 / OIDC / Rate Limiting
Node.js / Python
LLM Multi-Provider
GitHub Actions / Docker
Primary Data Store
Cache / Sessions / Pub-Sub
Object Storage / Assets
EC2 / RDS / S3 / CloudWatch / VPC
CI/CD Pipeline
Component Details
Frontend
Built with React and Next.js using the App Router with Server Components for optimal performance and SEO. Styled with Tailwind CSS. TypeScript for type safety across the entire frontend codebase.
Backend Services
Microservices architecture using Node.js and Python. RESTful APIs with OpenAPI specification. Event-driven communication for async operations. Rate limiting and request validation at the gateway.
AI Orchestration
Multi-provider LLM integration layer that routes requests to the appropriate AI model based on task type. Supports streaming responses, context window management, and prompt engineering pipelines.
Data Layer
PostgreSQL for relational data with strong consistency guarantees and automated migrations. Redis for session management, caching, rate limiting, and pub/sub messaging. Amazon S3 for durable object storage.
Security Layer
OAuth 2.0 / OIDC authentication with role-based access control. TLS 1.3 encryption in transit, AES-256 at rest. Automated dependency scanning and secrets detection in CI/CD.
Infrastructure
Deployed on AWS using EC2 for compute, RDS for managed PostgreSQL, S3 for storage, and CloudWatch for monitoring. Containerized workloads via Docker. VPC network isolation and security groups.
Want to dive deeper?
Read our full technical documentation for detailed architecture decisions, API design, and infrastructure configuration.
View Documentation