Summary
A backend engineering role focused on building intelligent recommendation platforms using machine learning models, scalable services, experimentation frameworks, and reliable production infrastructure.
Highlights
Build scalable AI-driven systems with ownership of production services, experimentation, and high-performance engineering.
Description
Job DescriptionWe are on the lookout for a Engineer II for our Core Recommendations Engineering squads.
This role is critical to building and operating the systems that power personalized discovery at scale.
You will own production services written in Go and Python that serve real-time recommendations across multiple surfaces, integrating tightly with machine-learning models and experimentation platforms.
You would be responsible for delivering, enhancing and maintaining the recommendation system which is robust, scalable and observable.
You will work closely with other engineers in the team on mainly backend engineering and few front-end related components.
You would collaborate with data engineering, data science, product partners to translate the recommendation strategies and models into production-grade engineering solutions.
Your squad is responsible for one of the foundational components of our customer discovery experience.
End-to-End Recommendation System: Design, build, and operate the full recommendation flow from candidate retrieval (recall-oriented systems, embeddings, and filters), through ranking (feature enrichment, model inference, scoring), to re-ranking (business rules, diversity, freshness, and policy constraints).
Ensure these pipelines meet strict latency, availability, and correctness requirements.
Production-Grade Recommendation Services: Build low-latency, high-throughput services in Go and Python that expose recommendation APIs, orchestrate model inference, and integrate with feature stores, vector databases, and downstream consumers.
System Ownership & Data Correctness: Own the lifecycle of recommendation systems, including data contracts, offline/online consistency, feature freshness, and backfills.
Ensure correctness, reproducibility, and debuggability of recommendation outcomes.
Operational Excellence at Scale: Operate Tier-1 recommendation services with strong observability, alerting, and incident response.
Perform post-mortems, manage error budgets, and continuously improve reliability, tail latency, and throughput.
Engineering Quality & Delivery Velocity: Follow strong engineering practices across Go and Python codebases, including testing strategies for ranking logic, CI/CD, safe rollouts, and performance regression detection.
Experimentation & Iteration: Enable fast iteration through A/B testing, shadow deployments, and controlled rollouts.
Partner with Product and Data Science to support model experimentation while maintaining system stability.
