Serverless changed backend engineering by removing much of the infrastructure work traditionally associated with running applications. Developers could deploy...
Microservices were supposed to make software easier to scale. Instead, many engineering teams discovered that they had scaled their architecture faster than...
GraphQL changed how developers think about APIs. Instead of maintaining multiple endpoints for different resources, clients could request exactly the fields...
Retrieval-Augmented Generation has become one of the standard architectures for building AI applications that need access to private, current, or domain...
LLM applications have a latency problem that traditional backend optimization cannot completely solve. A backend API can be carefully tuned, databases can be...
LLM APIs made it remarkably easy to add intelligence to a backend application. A developer can connect an API, send a prompt, receive a response, and move an...
There is a dangerous assumption around legacy software: if the code is old, the answer must be to replace it. That sounds logical until you look at what...
AI has moved from generating text to making decisions, calling APIs, updating records, and triggering workflows. For backend developers, that shift creates a...
“Move fast and break things” made sense when breaking things mostly meant fixing a buggy interface, rolling back a deployment, or patching an...
For years, the standard approach to building AI applications was simple. The application lived on the user’s device, but the intelligence lived somewhere...
