Getting Started

Welcome to Fennel

Fennel is a modern realtime feature engineering platform inspired by the ethos of PyTorch and is architected in the service of the following four design goals:

Fennel's Four Design Goals

  1. Easy to install, learn & use - be approachable to data science / ML teams and to make them fully self-dependent for end to end feature engineering workflows.
  2. Maintain data/feature quality via best practices - production ML systems often get plagued by data/feature quality issues. Prevention & detection of such issues (via best practices & tooling) should be a first-class citizen, not an afterthought. (Read more)
  3. Work equally well for batch & realtime - ML is transitioning from being predominantly batch oriented today to being more realtime. Fennel should support the exact same code working equally well for any combination of batch, streaming, & realtime computation.
  4. Keep cloud costs low - feature engineering often requires dealing with large volumes of data and gets very costly. For any workload, Fennel aims to be at least as cost-effective as a hand-rolled system while also being more powerful & easier to use. (Read more)

Use Cases

Fennel is a general purpose feature platform and is useful for a variety of use cases. Here are some of the most common ones:

  • Fraud Detection - relevant for fintech or anywhere money exchange is involved
  • Risk Engineering - identifying fake users, spam, bot rings and lot more
  • Credit Underwriting - use cases dealing with credit risk like credit cards, buy-now-pay-later, loan underwriting etc.
  • Insurance Underwriting - assessing the probability of filing a claim, expected damages etc.
  • Recommendations / Personalization - relevant for most marketplaces, e-commerce products, content feeds, notifications, email marketing campaigns etc.
  • Search Ranking - ordering available documents in response to a query based on relevance, personalization, or custom preferences
  • Marketing Personalization - choosing the right promotions/discounts to offer, the right time to send a campaign email, the right products to include in the email etc.

.. and lot more!

Getting Started With Fennel

Start here if you want to directly dive deep into an end-to-end example.

Or if you are not in a hurry, read about the main concepts first followed by some more details about Fennel's architecture.

We, the team behind Fennel, have thoroughly enjoyed building Fennel and hope learning and using Fennel brings as much delight to you as well!

On This Page

Edit this Page on Github