Skip to main content
Make your first Seer log and measure retrieval quality from unlabeled traffic.
Seer is in private beta. Email ben@seersearch.com to request an API key.
Prefer zero-friction integration? Use the @seer_trace decorator (Python) or wrapWithSeerTrace (TypeScript) to wrap your retrieval function and auto-log query+context.

​
Quick Start

1

Install the Seer SDK

pip install seer-sdk
2

Set your API key

Get an API key from the Seer Console and export it:
export SEER_API_KEY="seer_live_your_key_here"
3

Create quickstart.py

from seer import SeerClient

client = SeerClient()  # reads SEER_API_KEY from env

# Your retrieval step
def retrieve(query: str) -> list[dict]:
    # Replace with your real retriever
    return [
        {"text": "Christopher Nolan directed Inception.", "score": 0.95},
        {"text": "Nolan is British-American.", "score": 0.89}
    ]

query = "Who directed Inception and what is their nationality?"
context = retrieve(query)

client.log(
    task=query,                           # the user query
    context=context,                      # list of passage dicts
    metadata={
        "env": "prod",                    # environment tag
        "feature_flag": "retrieval-v1",   # for A/B testing
    },
)

print("Logged to Seer!")
# Events are auto-flushed when the process exits
4

Run it

python quickstart.py

​
What Happens Next

As logs arrive, Seer evaluates each retrieval and computes Recall, Precision, F1, and nDCG, all without labeled data. See Metrics for full definitions and worked examples.
By default, Seer evaluates ~10% of events (sample_rate defaults to server-side 10%). Set sample_rate=1.0 in your log() call to evaluate everything during testing. See Production Monitoring for recommended rates.
Use the Seer dashboard to:
  • Compare variants (change testing): filter by feature_flag to see A/B impacts.
  • Monitor production: filter by env or any metadata field.
If your app uses multiple retrieval steps (query decomposition, agent loops, parallel sources), the SDK auto-detects OTEL trace IDs to group related spans. See Multi-Hop Retrieval for patterns.

​
Next Steps