01 · Capture

Capture the execution evidence behind every outcome.

Inspect prompts, retrieval, tool calls, model responses, evaluator signals, latency, and token use in one trace model built for multi-step AI systems.

Trace Explorerproduction / tr_08fa71 200 OK
Resolve invoice exceptionrelease 2026.07.24 · prompt support-v18
gpt-5.62,841 tokens$0.0312.41s
Execution spans5 spans
agent.run
2.41s
retrieve.policy
184ms
model.reason
1.36s
tool.customer_record
421ms
model.response
386ms
model.reasonScore 0.91
system

Apply the current billing policy. Use tools only when the customer record is required.

Modelgpt-5.6
Promptsupport-v18
Input1,984 tok
Output312 tok
Groundedness0.91
Policy adherence0.78
STAGE OUTPUTS

Leave this stage with evidence the next one can use.

Each stage is configured around the customer's application, risk, quality standard, operational ownership, and release policy.

01

Nested execution

Move from session to trace to span without losing prompt, model, or tool context.

02

Version-aware evidence

Link every span to the prompt, model, configuration, and deployment that produced it.

03

Operational search

Filter by environment, release, feature, customer segment, failure mode, latency, score, or cost.

04

Failure reconstruction

Preserve errors, retries, tool results, retrieval evidence, and evaluator explanations for review.

CONNECTED RELEASE RECORD

No stage operates in isolation.

The evidence links traces, versions, evaluation runs, release gates, production monitors, and cost attribution.

Inspect architecture →
ENGINEERING NOTES

Practical thinking for production AI quality.

View the research library →
INSTRUMENT ONE REAL FLOW

Start with the traces behind a production outcome.

Map the execution path, identify missing context, and agree the trace evidence required before evaluation begins.

Scope trace coverage