engine room / answer trace

wrun_01M5TQVFQ6ZS1TEJC7DR70PDGC

Answered in 16.48s, first token at 12.91s, 2,137 tokens, $0.00033, 1 verified citation.

Visitor identity is a one-way hash; message content never enters traces.

Timeline

Queue

enqueue + dispatch

Guardrail

allow

Dispatch

step handoff

Retrieval

hybrid

Dispatch

step handoff

Generation

glm-5.3-flash

2.81s
2.82s
1.09s
469ms
4.00s
5.04s
0first token 12.91s16.21s

The gray spans are durable-workflow dispatch: each stage is an independently retried, checkpointed step, and the handoffs are what that crash-safety costs.

What happened

  1. Admitted at the edge

    A 18-character message from caller e43d0d0a… passed schema validation, the per-caller rate limit, and both token budgets (charged an estimate of 1,029 tokens, settled later against actual usage).

  2. Became a durable workflow

    The endpoint never calls a model. At 23:55:37.844 the request was enqueued as run wrun_01M5TQVFQ6ZS1…; tokens stream into the run’s persistent log, which is why a page reload can reattach mid-answer.

  3. Cleared the guardrail

    gpt-oss-safeguard classified the message against policy 6bb17bbe0908 in 2.82s: allow.

    “Valid question about Joshua Bharathi's location”

  4. Pulled its evidence

    Hybrid search (dense embeddings fused with BM25) returned 6 passages in 469ms; only these entered the model’s context.

    1. faq#00.0328
    2. profile/about#00.0167
    3. profile/experience#00.0164
    4. profile/about#10.0159
    5. profile/journey#00.0156
    6. profile/awards#20.0154

  5. Generated

    zai/glm-5.3-flash streamed the answer, first token 12.91s in, spending 2,109 tokens in / 28 out ($0.00033).

  6. Verified and settled

    Every citation was checked against the concept registry and resolves. Token budgets were settled against actual usage and this trace was written.

    faq

Record

Status
finished (stop)
Admission
admitted
Guardrail reason
Valid question about Joshua Bharathi's location
Tokens in / out
2,109 / 28
Cost
$0.00033
Model
zai/glm-5.3-flash
Prompt version
45b79cdd7ce5
Policy version
6bb17bbe0908
Input size
18 chars
Caller (hashed)
e43d0d0add5c896e
Request id
e5cb80cf-d645-45c7-88a7-34825cf23658
Enqueued
2026-10-25T23:55:37.844Z
Finished
2026-10-25T23:55:54.055Z