LLM

Asking a model to 'reply in JSON' and hoping is how you end up parsing prose at 2am. Modern models expose schema-constrained output and tool calling that guarantee the shape — the trick is choosing the right one and validating anyway.

• AI
• LLM
• TypeScript

An LLM only knows what it was trained on and what you paste into the prompt. RAG is how you give it your own knowledge — your docs, your data — without retraining anything. Here's the pipeline, end to end.

• AI
• RAG
• LLM