❓ Quiz — advanced/session-08

Q1 — Nature of the API and memory

A developer is surprised: “I sent the client file to the model three requests ago, why doesn’t it remember it anymore? » What is the right architect’s explanation?

HAS The model has a 30 minute memory; beyond that, the data must be returned B The API is stateless: the model only sees what is returned in the current request; history management is entirely the responsibility of the application C You must activate the option memory: true in the query parameters D Prompt caching would have preserved the file; it is its absence that causes forgetting

Q2 — Saturation arithmetic

Window of 200,000 tokens ⚠. Fixed costs (system + tool definitions): 7,000 tokens. Exit reserve: 8,000 tokens. Each full spin (exchange + raw tool result) adds ~3,300 tokens to the history. Order of magnitude of the saturation turn?

HAS ~25 rounds B ~56 rounds C ~150 rounds D No saturation: old tours are automatically eliminated by the API

Q3 — Choice of context strategy

A long-term investigation agent must both remember decisions made 100 turns ago and reason carefully about the last 5 exchanges (exact tool results included). What strategy?

HAS Strict sliding window on the last 20 revolutions: simple and without compression cost B Full summary of all history at each turn: maximum compactness C Hybrid: cumulative summary of old turns + last N turns kept verbatim D Increase max_tokens to enlarge the context window

Q4 — Reliability of the context summary

In a digest-compression architecture, which statement is correct?

HAS The summary is lossless if we use the same model as the agent B The summary is a model output: lossy compression, possible hallucination; in a regulated context, the full history must be kept out of context (audit log) C At temperature 0, the summary is guaranteed to be faithful to the history D The summary must include all raw tool results to remain usable

Q5 — PostToolUse hook and context pollution

A tool search_transactions returns 40 fields per transaction and up to 10,000 rows. The agent uses 4 fields and a few flagged lines. What is the best practice?

HAS Insert everything in context: the more information the model has, the better it reasons B A PostToolUse hook that logs the full raw result (audit), then inserts only relevant fields into context with a row cap C Delete the raw result and only log the filtered version, to save storage D Ask the model, in the prompt, to ignore irrelevant fields

Q6 — Scratchpad pattern

What is the decisive advantage of the boss? investigation-scratchpad.md » (persistent external memory) compared to the summary in context alone?

HAS It makes the summary in context useless in all cases B It guarantees the accuracy of the facts noted, because the file is not generated by the model C It provides selective memory that survives compressions, restarts and resets, inspectable by a human and transferable to another agent D It is automatically synchronized with the context window by the API

Q7 — Verification of provenance

A generated report cites {"tool": "check_sanctions", "call_id": "call_0042", "record_id": "SANC-2211-08"}. What should the application do before considering the assertion as sourced?

HAS Nothing: a structured quote produced at temperature 0 is reliable B Check that the format of the record_id respects the expected regular expression C Resolve pointer against audit log: the call_id exists, and the record_id appears well in the raw result of this call D Ask the model again to confirm their quote in a second call

Q8 — Reproducibility

For a regulatory report, management requests the guarantee of being able to “regenerate exactly the same report” from the same inputs. Correct response from the architect?

HAS Yes: temperature 0 + pinned model version + seed ⚠ guarantee identical bit-to-bit outputs B No, and this is why generative AI is unusable in a regulated sector C Regeneration is quasi-deterministic but not guaranteed bit by bit (inference infrastructure); the guarantee of conformity is based on the exhaustive audit log of the output actually produced and all its context D Yes, provided you use the Batches API, which is deterministic by design

Q9 — API batches: analysis

Your batch of 80,000 requests displays processing_status: "ended". Which statement is correct?

HAS All requests were successful; results are in order of submission B Each request has an individual status (succeeded , errored , canceled , expired ) and the order of the results is not guaranteed: the counting is done query by query, correlated by custom_id C Error requests were automatically resubmitted by the API D Results are pushed by webhook; no counting is necessary

Q10 — Economical batch + cache optimization

Non-interactive nightly reprocessing of 80,000 files sharing a system prompt of 3,200 tokens. Which configuration minimizes the cost, and why?

HAS Synchronous calls with cache: cache is incompatible with batches B API batches only: the 50% reduction ⚠ makes the cache irrelevant C API batches with the system block in cache_control : the batch reduction of 50% ⚠ (input and output) combines with the reduced rate of cache reads on the shared prefix D Parallelized synchronous calls: batch is more expensive because it includes the 24-hour SLA ⚠

Quiz — Advanced Level, Session 8

“Context, reliability & provenance »