How nasa-impact/veda-backend works
Project type and purpose
• This is a full-stack, serverless backend for a SpatioTemporal Asset Catalog (STAC).
• It uses AWS CDK to deploy a VPC, RDS (PgSTAC), serverless APIs (STAC API, TiTiler raster API, ingest API), a DynamoDB queue, and an S3 website.
File structure
• Top-level CDK constructs under directories:
• network/infrastructure (VPC, subnets, gateways)
• database/infrastructure (RDS with pgstac)
• ingest_api/infrastructure (API Lambda + DynamoDB)
• stac_api/infrastructure (STAC API Lambda + API Gateway)
• raster_api/infrastructure (TiTiler Lambda + API Gateway)
• s3_website/infrastructure (static site bucket + CloudFront or S3 website)
• permissions_boundary, standalone_base_infrastructure (shared infra)
• Runtimes for each service under their respective runtime/ directories.
• Common utilities (common/auth) and scripts for local testing, deployment, and maintenance.
README for additional insights
• Key technologies: stac-fastapi, pgstac, titiler, titiler-pgstac, AWS Lambda, API Gateway, CDK.