quiz bu lëkkale ak tegtal · Diggante Niveau
Quiz Ingénieur Contexte
Dafay saytu ni ingenieur context di wuutee ak laaj, lu tax context nekk budget token, ak pexe yu melni chargement ci waxtu wi war ak cache ci saa si.
Laaj 1 de 8
Lan moo gëna wuutale ingenieur contexte ak ingenieur prompt?
Xoolaatal laaj bu nekk8 laaj
Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.
- Lan moo gëna wuutale ingenieur contexte ak ingenieur prompt?
- Dafay wax ci kàdduy laaju jëfandikukat bi kese
- Dafay muur lépp lu model bi di gis, lu ci melni done yiñ jëlee, njariñu jumtukaay bi ak mémoire bi, du ci kàddu yi kese
- Dafay laaj nga defar bu baax diisaayu model bi
- Dafay wax ci xeetu defar nataal kese
- Lan la gëstub 'réer ci digg bi' digle ci wàllu dugal xaalis bu gudd?
- Modèle yi dañuy jàng wàll wu nekk ci dugal bu gudd
- Model yi duñu bàyyi xel ci ndoorte li ñuy dugal
- Modèle yi dañuy faral di jëfandikoo leeral yi ci ndoorte ak ci njeextelu dugal bu gudd, moo gën leeral yi ci digg bi
- Model yi mënu ñu def lu ëpp 1000 jeton
- Lan moo waral guide bi dafay wax ni budget la?
- Bépp token dafay njëg xaalis ak latency, te mbir yu amul njariñ mën nañu yàq model bi
- Fournisseur yi dañuy sàkku xaalis weer wu nekk ngir contexte
- Model yi duñu nangu dugal xaalis bu ëpp benn dolaar
- The word refers only to the compute used in training
- Lu la just-in-time loading?
- Loading the entire codebase into the window at the start of a session
- Retraining the model whenever new data arrives
- Caching every model output permanently
- Giving the agent lightweight references, such as file names, and letting it fetch full content with tools when needed
- Lu tax mu am solo ngir def stable content, ni system instructions ak tool definitions, ci tambali ci anam gu amul benn?
- Dafay tax model bi gëna am gis-gis
- Every API requires this order
- An identical prefix can be reused through prompt caching, which lowers cost and latency
- It hides the instructions from the user
- Which approach to tool design does the guide recommend?
- Tools should return all raw data so nothing is missed
- Tools should return concise, high-signal output and support filtering or pagination
- Many tools with overlapping descriptions give the model more choice
- Tool descriptions should be left empty to save tokens
- Which statement is a misconception the guide warns about?
- Large context windows increase cost per request
- Tool definitions take up tokens on every turn
- Retrieval can reduce how much text goes into the window
- A bigger context window removes the need to curate what goes in
- What role do sub-agents play in context engineering?
- They explore in their own separate context and return condensed findings to the main agent
- They double the context window of the main agent
- They replace the system prompt
- They store user passwords between sessions