Summary
What you’ll impact
Our company is seeking an innovative Machine Learning Engineer to join the Algorithm team in New York, reporting to the Global ML Lead. The role focuses on 0→1 development, building, scaling, and optimizing feed ranking and personalization systems, including designing features, architecture, and A/B testing. The engineer will collaborate with cross‑functional teams to create a healthy content ecosystem.
Responsibilities
What you'll do
- Design and build the ranking stack end to end, from candidate retrieval, pre-ranking, final ranking, to the assembly layer that decides what the feed actually looks like, and own it in production.
- Turn raw signals into features: user behavior, context, and the outputs of the content-understanding models we run on every post at upload time.
- Work with our backend and infra engineers on serving: feature stores, inference, latency budgets.
- Partner closely with cross functional teams (product, design, operations) to build innovative experiences
- Design and execute A/B testing framework to evaluate new algorithms. Establish the baseline metrics and iteratively ship new models to drive immediate and measurable impact.
- Balance multiple, sometimes competing, ranking objectives such as: relevance, content diversity, freshness, and creator fairness, to ensure a healthy product ecosystem.
- Set the engineering bar as the team grows, ensuring high quality code reviews, testing, and AI usage.
Requirements
What you’ll bring
- A bachelor’s or higher in computer science, machine learning, statistics, math, or a similarly quantitative field.
- 5+ years in industry as a machine learning engineer or applied researcher, having led and shipped multiple models & products
- Proven hands-on experience building and deploying Recommender Systems, Feed Ranking, Search Ranking, or relevant models. Experiences working at a social discovery platform is a plus.
- Strong proficiency in Python and ML frameworks (PyTorch, SFT techniques, etc) and an understanding of how to serve these models with low latency.
- Comfortable with large scale data work such as Spark or Flink
- A strong understanding of consumer social dynamics, network effects, and UGC is preferred.
- Thrive in ambiguous, fast-paced environments, enjoy building end-to-end systems, and are biased towards action and product-attuned building.