Beginner
Foundations

What you are actually calling

The model behind the API: what it reads, what it returns, what it costs and where it stops.

What you will be able to do

Read a provider’s pricing page and a response object and know what both are telling you.

Prerequisites
Ordinary software engineering. No machine learning.
Length
8 lessons · 39 min
Start course

Syllabus

  1. 1What is an AI Engineer?A single support-reply feature, taken through the failures it hits in production: supplying a refund policy, grounding replies in help-center articles, structuring output for a ticket tool, and handling prompt injection in customer email. The role is the engineering around the model call.5 min
  2. 2What AI engineers actually do all dayThe day-to-day work is context plumbing, schema design, evaluation and cost control rather than model training. The prerequisite that carries most of the weight is ordinary software engineering.7 min
  3. 3Large Language Models (LLMs)What goes into a request, what the model holds while answering, and what comes back. The context window is a shared budget for input and output, and some model limits can be fixed while others have to be designed around.4 min
  4. 4Stop estimating tokensCharacter and word counts disagree with real token counts, which is a problem because every limit you hit is counted in tokens. How byte pair encoding splits text, and how to budget a request so a response is not cut off mid-sentence.6 min
  5. 5Context WindowInput and output come out of one budget, so a request that fits can still fail partway through writing. Cached tokens change what you pay, not how much room they take.4 min
  6. 6InferenceWhat happens between your request and the answer: time to first token against total response time, why output length drives latency, where token costs accumulate, and the inference failures to handle in code.5 min
  7. 7Sampling ParametersTemperature and top-p change how a token is drawn, not what the model knows. And temperature 0 is not a reproducibility guarantee.4 min
  8. 8Prompt CachingA cache read costs a tenth of base input on most Claude models, and a date in the system prompt switches the whole thing off. Below the per-model minimum, caching is skipped and no error is returned.4 min