Architect, Real-Time ML Recommender Platform

ByLabs; San Francisco (CA); On-site; USD 180,000 - 280,000; Full time

Job summary

ByLabs is seeking a Principal Recommendation System/Platform Engineer to upgrade our core recommendation architecture, transforming a rule-based personalization system into an industrial-scale, real-time ML platform supporting trading products, content feeds, the community, and AI-powered investment workflows.

You will partner with data science, growth, and operations to deliver a high-availability foundation for personalization at scale, serving 80 million users globally.

Qualifications

• At least 4 years of recommender systems engineering experience at a consumer-facing internet company and hands-on production issues experience.
• Strong CS fundamentals and proficiency in Go, Java, or C++ with well-structured code.
• Familiarity with PyTorch or TensorFlow and end-to-end workflow of training, exporting, deploying, and serving models.

Responsibilities

• Develop high-throughput, low-latency recommendation services and ensure reliability with degradation and disaster-recovery mechanisms.
• Build and optimize large-scale model training infrastructure, incl. distributed training and job scheduling.
• Develop deployment and online inference services with autoscaling and canary releases for high traffic.
• Establish feedback loop from offline eval to production monitoring and incident diagnosis.
• Collaborate with teams to standardize training, deployment, and inference across scenarios.

Skills

Recommender systems

Go

Java

C++

PyTorch

TensorFlow

Tools

PyTorch

TensorFlow

Job description

ByLabs is seeking a Principal Recommendation System/Platform Engineer to upgrade our core recommendation architecture, transforming a rule-based personalization system into an industrial-scale, real-time ML platform supporting trading products, content feeds, the community, and AI-powered investment workflows.

You will partner with data science, growth, and operations to deliver a high-availability foundation for personalization at scale, serving 80 million users globally.