Learn / Subjects / AI Building with LLM APIs For developers putting models into real software: tokens and cost, structured output, tool use, retrieval, evals, and agents you can trust in production.
55 bytes · Taught in 1 course
Every byte Code and Build with AI › Ch 4 · Build Your First AI Feature 1 What changes when your code calls a model Learn 2 Tokens are your unit of cost and speed Learn 3 Pay once for the part that never changes Try This 4 Why temperature 0 won't make a model repeat itself Unlearn 5 Make a model return JSON that fits your schema Try This All 21 here › Code and Build with AI › Ch 6 · Choose the Right Model for Your Work 1 Put each candidate AI model through your eval cases Try This 2 Why a pricier AI model can cost less per finished job Versus 3 Size a local AI model to your machine's memory Try This 4 Why your local AI model forgets the file it just read Learn 5 Find what your AI feature lacks before fine-tuning it Scenario All 7 here › Code and Build with AI › Ch 7 · See Inside Every AI Request You Send 1 Count an AI prompt's tokens on the model you'll use Try This 2 Why an AI prompt costs more outside English Learn 3 Work out what a screenshot costs your AI feature Try This 4 Each tool you add costs your AI agent before it works Versus 5 What really runs when you connect an AI tool server Learn All 21 here › 
