Retrieval and embeddings · Lesson 12 of 16 · 7 min

Review metadata filters

Metadata filters are part of retrieval correctness and access control. A relevant passage from the wrong tenant is not an acceptable result.

Open the lab Course map

Mastery 0/3Run itExplain itApply it

learner@clairos:~ $

↑ history · Tab complete · Ctrl+C cancel · Esc, Tab to leaveExit 0

Your objectives

Hints are always available. Wrong turns never cost XP.

1
Read the case evidence
Read /opt/ai-lab/cases/aiops-retrieval-embeddings-012.json. Separate the recorded facts from your assumptions.
Hint & worked example
Metadata filters are part of retrieval correctness and access control. A relevant passage from the wrong tenant is not an acceptable result.
The complete case JSON is read successfully.
```
cat /opt/ai-lab/cases/aiops-retrieval-embeddings-012.json
```
Common mistake to avoid: Relevance is more important than the tenant boundary.
2
Isolate the deciding signal
Select the result_tenant field from the case JSON. What does it establish—and what does it not establish?
Hint & worked example
Reject the cross-tenant result even though its text is relevant.
A successful JSON query isolates result_tenant.
```
jq .result_tenant /opt/ai-lab/cases/aiops-retrieval-embeddings-012.json
```
Common mistake to avoid: Using a different case or an unrelated signal as the justification.
3
Record your operational decision
Choose the justified code: enforce-tenant-filter, change-without-evidence, or ignore-the-signal. Save that one code followed by a newline in /home/learner/lab/aiops-retrieval-embeddings-012-decision.txt.
Hint & worked example
Reject the cross-tenant result even though its text is relevant.
The decision artifact contains the correct code, not just a matching command in history.
```
printf '%s\n' 'enforce-tenant-filter' > /home/learner/lab/aiops-retrieval-embeddings-012-decision.txt
```
Common mistake to avoid: Relevance is more important than the tenant boundary.

Understand the pattern

Metadata filters are part of retrieval correctness and access control. A relevant passage from the wrong tenant is not an acceptable result.

Interpret the evidence

Reject the cross-tenant result even though its text is relevant.

A tempting mistake

Relevance is more important than the tenant boundary.

All names, measurements, prices, versions, and thresholds in this exercise are fictional training data. The evidence supports this bounded scenario, not a claim about a live provider.

Worked example and expected result

$ cat /opt/ai-lab/cases/aiops-retrieval-embeddings-012.json

{
  "fixture": "Fictional training data; not a live measurement.",
  "request_tenant": "a",
  "result_tenant": "b",
  "text_relevant": true
}

Exit status: 0. The command completed successfully in the training fixture.

$ jq .result_tenant /opt/ai-lab/cases/aiops-retrieval-embeddings-012.json

"b"

Exit status: 0. The command completed successfully in the training fixture.

$ printf '%s\n' 'enforce-tenant-filter' > /home/learner/lab/aiops-retrieval-embeddings-012-decision.txt

(no output; exit 0)

Exit status: 0. The command completed successfully in the training fixture.

Try a variation

Name the assumption that would have to change for “enforce-tenant-filter” to stop being the justified action. Do not silently replace missing evidence with a guess.

Explain it

What does the evidence in “Review metadata filters” justify?Reject the cross-tenant result even though its text is relevant.Relevance is more important than the tenant boundary.Make the change now and gather the missing evidence afterward.

Apply it elsewhere

A colleague concludes: “Relevance is more important than the tenant boundary.” What should you do?Accept it because the training command completed.Treat this fixture as a live measurement.Reject that conclusion; return to the case evidence and its stated limits.

Keep the momentum.

Complete the lab, then explain and apply the pattern.

Next: Compare retrieval configurations Try an incident drill

Anonymous lesson and mastery progress lasts for this session. Sign in to keep synced lesson XP between sessions; mastery stays device-local. No speed bonus, lost lives, or hint penalties.

Keep learning

Case methodology: evidence, limits, decision