Speaker
Description
While many programming languages have improved their dependency handling to simplify collaboration, real software projects include multiple interacting components: backend, frontend, database, and worker services. Manually managing connection strings and database instances quickly becomes a significant bottleneck.
This talk explores how Aspire helps solve this issue. With Aspire, the entire developer environment is configured as code, featuring connectors for databases (relational, document, and vector), polyglot applications (Python, .NET, and TypeScript), and AI providers (OpenAI, Ollama). Each integration comes with default setups for service discovery, observability, health checks, and resiliency. Because the orchestration is defined as code, it is easily version-controlled and shared; onboarding a new collaborator becomes as simple as cloning the repository and running the Aspire project.
In my experience, Aspire significantly lowers the barrier for new collaborators. Furthermore, the telemetry provided by its observability extensions makes a tangible difference in project transparency. To demonstrate this, I will showcase a RAG (Retrieval-Augmented Generation) application built with a local LLM, a vector database, and a decoupled frontend/backend. I will highlight how Aspire’s telemetry helped me diagnose and fix hidden issues in the data flow that would otherwise have been difficult to track.