cleandataengineering.aiAgent Harness, drawn simply
You are here: Tech stack › 18 · Knowledge & retrieval

Agent Harness Tutorial · Tech stack

18 · Knowledge & retrieval

Finds sourced passages, filtered by what this user may see. Family: Think · one turn, in order.

AZURE OPTIONSAzure AI Search· considered: fallback if the corpus outgrows pgvectorContent Understanding· consideredOUR CHOICEpgvector on the same PostgreSQLnot built yet: phase 2OPEN-SOURCE OPTIONSpgvector◐ planned: chosenQdrant on Container Apps· considered: alternativeWHO DOES THE WORKAzure gives 45% · we code once 30% · each agent team still does 25%
Dark boxes are what we run. Solid boxes are used or planned; dashed boxes were considered or dropped (crossed). The bar: grey is what Azure gives, dark is what we code once, hatched is what each agent team writes.
Why our choiceOne store, one backup, one access model; the entitlement check is just a SQL filter.
OptionWhere fromVerdictWhy
pgvector on the same PostgreSQLours✓ chosennot built yet: phase 2
Azure AI SearchAzure· consideredfallback if the corpus outgrows pgvector
Content UnderstandingAzure· considered
pgvectoropen source◐ plannedchosen
Qdrant on Container Appsopen source· consideredalternative

Not built yet in the claims assistant; designed in the Series.

Compared: Where do knowledge passages live and get searched?

pgvector in Azure PostgreS…open sourcechosenAzure AI SearchAzurelaterQdrant on Container Appsopen sourcealternativePineconevendornot evaluated
Dark: what we chose. Crossed and dashed: rejected or not for us. From the Series.
OptionFromVerdictWhere it runsHybrid keyword + vectorEntitlement filterExtra infrastructureWhen to pick
pgvector in Azure PostgreSQLopen sourcechoseninside our PostgreSQLyes, with Postgres full texta SQL filter before rankingnone: one store, one backupdefault
Azure AI SearchAzurelatermanaged search serviceyes, plus semantic rankerfilters on indexed fieldsa search servicecorpus outgrows Postgres
Qdrant on Container Appsopen sourcealternativea container we runyes, dense plus sparsepayload filtersa service plus its storagededicated vector store needed
Pineconevendornot evaluatedmanaged SaaSyes, sparse-densemetadata filtersa SaaS outside Azurenamed only
DecisionPassages live in pgvector in the same PostgreSQL, so entitlement is a SQL filter applied before ranking and there is one store to back up. We would switch to Azure AI Search when the corpus outgrows Postgres or we need its semantic ranker.