Before discussing context windows, read this short excerpt and notice what information the model would need to answer well.
Build AI-supported lessons that keep the reading in charge.
Twigby turns course material into focused quests where people read, respond, and ask contextual AI questions without leaving the lesson.
Back
Strengths and Weaknesses of LLMs
AIReflect
Where would a long-context model change the answer?
Write a short answer. Twigby responds with feedback tied to this point in the quest.
A model can compare the current question with the assigned reading, the prompt, and the previous response.
Quest lifecycle
One workspace for the whole lesson path.
- 1
Build the quest
Arrange reading, media, checks, debate, and visualization blocks into one lesson flow.
- 2
Run the lesson
Participants move through the same focused reader, with prompts and progress attached to the quest.
- 3
Ask in context
AI support opens beside the current moment and carries the lesson context into each answer.
Back
Strengths and Weaknesses of LLMs
AIBefore discussing context windows, read this short excerpt and notice what information the model would need to answer well.
Reflect
Where would a long-context model change the answer?
Write a short answer. Twigby responds with feedback tied to this point in the quest.
A model can compare the current question with the assigned reading, the prompt, and the previous response.