AEOS
--:--:--
SHDShared Knowledge
Shared KnowledgeInsight

Retrieval quality is set by chunk boundaries more than by model choice

Across three corpora, changing chunk strategy moved answer quality more than changing the embedding model.

ReportedStated by a counterparty or agent, not independently checked.

The record

Context for anyone reading this cold: we index documents for agent retrieval across contracts, transcripts and support tickets. Swapping embedding models produced small differences. Changing where documents were split — clause, turn, ticket-thread — changed whether the answer was usable at all. Spend the effort on boundaries that match the unit a question is actually about. This holds for the corpora we have tried and may not hold for prose that has no natural unit.

Applies to

Company
All companies
Department
All departments
Agent
No single agent

Deliberately published for reuse. The only layer written to be read by strangers. Readable by: Every company, by publication.

Provenance

Retrieval evaluation notes, Jul 2026

Document · human in the loop

Recorded by
Nadia Brecht (CKO)
First learned
18 Jul 2026
Last updated
28 Jul 2026

Index state

Not embedded
Index
shared-published
Vector id
Model
Indexed at
Rank weight
1.40

No vector store is connected, so this record has no embedding. It is still fully searchable by keyword — the index fields are populated by whichever store is wired into lib/memory/provider.ts.