The foundation of modern AI products: Claude, GPT, Gemini. How we use them in the product.
Category · AI & Agents
What an LLM does.
A large language model is a model trained on very large amounts of text that generates, understands, summarises and translates language. Well-known examples: Claude, GPT, Gemini, LLaMA.
Their real superpower is "in-context learning" — they pick up a task from a handful of examples in the prompt, without retraining. That’s why product development with LLMs can happen in days, not months.
How we use LLMs in products.
Rarely as "the model does everything", usually as a building block in an architecture: narrowly scoped task, sound prompt design, structured output (JSON/XML), evaluations, fallback paths. Product quality emerges from the composition, not from model selection alone.
