Predict customer behavior the speedy way
When it's time to put down your notebooks
- Data ingress and integrations
- Fully prepared consumer data
- Identity resolution
- Algorithm tuning
- Feature engineering
- Validation & reporting
- Probability calibration for useful scores
- Geonormalization
- Explainability
- Bias detection & mitigation
- Real-time and batch inference
- SOC-2, CCPA, and other regulation
- Lifecycle management
Faraday is ready
Everything on the leftabove is included. Just …
- Connect to your existing data sources
- Declare your prediction objectives
- Review automatic reporting
- Deploy your predictions anywhere
Which behavior will you predict?
Let's speedrunwith the API
Create Connections to your data sources
Faraday supports data warehouses like Snowflake and BigQuery, databases like Postgres, and cloud buckets like S3. Or just start by uploading CSV with POST /uploads
.
Create Cohorts to represent key groups
Declare your prediction objective
Faraday has built-in objectives for key customer behaviors. This template uses an outcome to make the necessary predictions.
Use a Scope to make your predictions
Scopes let you choose a population (using Cohorts) and a payload, including the objective you just created, to prepare for deployment.
Use a Target to deploy your predictions
You can add a target to your pipeline for every deployment destination you need for your use case.
Ready to ship?
Skip the ML struggle and focus on your downstream application. We have built-in sample data so you can get started without sharing yours.