Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.neuronsearchlab.com/llms.txt

Use this file to discover all available pages before exploring further.

The pipeline config page gives you control over the multi-stage recommendation pipeline. Each recommendation request flows through a series of stages, and you can enable, disable, or tune each one independently.

​
Pipeline stages

Recommendations are produced through these stages, in order:
StagePurposeKey config
Candidate generationRetrieve initial candidates from the vector index using cosine similaritycandidate_limit — max candidates to pull
Feature enrichmentAttach metadata and computed features to each candidate—
ScoringRun the ML model scoring pass over enriched candidates—
RulesApply business rules such as boost, bury, pin, filter, cap, diversity, dedupe, reorder, and grouping controlsCan be disabled entirely
RankingFinal sort by adjusted scores—
Post-processingTrim to page size and apply any final transformspage_size — number of items returned

​
Configuring the pipeline

1

Open Pipeline Config

Navigate to Console > Pipeline Config.
2

Review each stage

Each stage is shown as a card with its current status (enabled/disabled) and configuration.
3

Toggle stages

Toggle a stage on or off. Disabled stages are skipped entirely at runtime.
4

Tune stage settings

Where available, adjust stage-specific settings like candidate_limit or page_size.
5

Save changes

Save your changes.

​
When to disable stages

  • Disable Rules to produce pure ML-ranked results. Useful as a control arm in an A/B experiment to measure whether your business rules help or hurt engagement.
  • Disable Feature Enrichment if you are testing raw embedding similarity without metadata features.

​
Tuning candidate limits

The candidate_limit in the candidate generation stage controls how many items are pulled from the vector index before scoring. A higher limit gives the scoring model more candidates to choose from but increases latency. Start with 100-200 and increase if you notice the final results lack diversity.

​
Tuning page size

The page_size in post-processing determines how many items are returned in the API response. This does not affect internal processing — all stages operate on the full candidate set. The trim happens at the end.

​
Pipeline and experiments

When running an A/B experiment, each variant can reference a different pipeline config. For example:
  • Control: default pipeline with all stages enabled
  • Treatment: pipeline with the rules stage disabled
This lets you isolate the impact of specific pipeline stages on recommendation quality. See A/B Testing for setup details.

​
Fallback behaviour

If no pipeline config exists for your team, the system uses sensible defaults:
  • All stages enabled
  • candidate_limit: 100
  • page_size: 20
You only need to configure the pipeline if you want to change these defaults.