AI Discovery Feed Algorithm Engineer

Intevo AI

Remote · $200k+ / year · Full-time · Posted 5 months ago

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

About the role

The discovery feed is the single most important surface in our marketplace — it drives 48% of all buyer-seller connections and directly influences $340M in annual GMV. The feed algorithm decides what 2.8 million users see when they open the app, and small improvements translate into measurable revenue impact. A 1% lift in feed engagement last quarter generated $4.2M in incremental GMV. We're hiring a senior engineer to own the ranking model and its supporting infrastructure. The current system uses a two-tower architecture for candidate retrieval and a gradient-boosted ranker for final ordering. Your job is to evolve both stages: exploring transformer-based architectures for retrieval, adding real-time behavioral signals to ranking, and building the experimentation framework that lets us test changes safely on a marketplace where both sides of the transaction are affected. You'll work with a product manager, a data scientist, and two backend engineers who handle serving infrastructure. This is a high-leverage IC role where your technical choices have direct business impact measurable in dollars per week.

Fully remote.

Responsibilities

Own the discovery feed ranking model end-to-end: training, evaluation, and serving
Evolve candidate retrieval from two-tower to more expressive architectures
Incorporate real-time behavioral signals into ranking features
Build experimentation frameworks for safe A/B testing on a two-sided marketplace
Analyze feed performance metrics and identify optimization opportunities
Partner with product to translate business goals into model objectives

Requirements

7+ years in recommendation systems or ranking at scale
Deep expertise in learning-to-rank, retrieval models, and online experimentation
Production experience serving models at millions of daily active users
Strong Python, PyTorch, and SQL skills
Track record of measurable business impact from model improvements