fix(captain-memory): raise RecallService timeout 0.5s -> 2.0s
Real-world observation: OpenAI embedding API takes 200-400ms typical, plus pgvector query overhead, the 500ms budget was being exceeded frequently, silently dropping memory recall. Agent typing delay is already 2-15s humanized, so a 2s recall budget is well within UX tolerance and gives ~4-5x margin over typical embedding latency. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
parent
1ce07cc78c
commit
5874029a03
@ -1,5 +1,5 @@
|
||||
class Captain::ContactMemories::RecallService
|
||||
TIMEOUT_SECONDS = 0.5
|
||||
TIMEOUT_SECONDS = 2.0
|
||||
DEFAULT_TOP_K = 5
|
||||
|
||||
def initialize(contact:, query_text:, unit_id: nil, top_k: DEFAULT_TOP_K)
|
||||
|
||||
Loading…
Reference in New Issue
Block a user