Staff Machine Learning Engineer - Moloco Commerce Media

Moloco•Menlo Park, CA

13h•Onsite

About The Position

Moloco builds some of the most powerful AI advertising solutions in the world. Our name—short for "machine learning company”—reflects our core mission: democratizing access to the advanced AI that has historically been reserved for tech giants. Led by machine learning pioneers who built some of the most successful ad systems at Google, including YouTube's monetization engine and key search advertising technologies, we're transforming how businesses grow and compete in the digital economy. Built with AI from day one, Moloco’s planet-scale machine learning platform powers a suite of solutions for advertising growth and monetization. Moloco Ads is an AI-powered platform that delivers real business outcomes for mobile app marketers through performance-based user acquisition. Moloco Commerce Media enables retailers and marketplaces to build revenue-generating ad businesses that balance user experience and advertiser performance. Moloco is headquartered in Silicon Valley, with offices in Seattle, New York, San Francisco, Seoul, Beijing, Singapore, Gurgaon, Tokyo, Shanghai, London, Tel Aviv, and Berlin. Moloco is a truly rewarding place to work and in an exciting period of growth, which you could be a part of. Join us today and apply now! The Impact You’ll Be Contributing to Moloco Design and deliver cutting-edge machine learning models that power Moloco’s Commerce Media products, serving as the core “engine” behind ads ranking, recommendations and search for retail media customers. Advance the performance, scalability, and reliability of our core ad prediction and retrieval systems so customers can unlock meaningful business outcomes from their first campaigns onward.

Requirements

8-12 years of experience in applied machine learning or machine learning engineering, with a track record of shipping production models in ads, recommender systems, search, or large-scale ecommerce.
Deep hands-on expertise with modern ML toolchains (e.g., Python, PyTorch/TensorFlow/JAX, feature stores, distributed training, and large-scale data processing frameworks) to solve complex ranking and prediction problems.
Experience building and optimizing models such as CTR predictors, retrieval models, conversion models, or recommendation systems, ideally including exposure to transformer-based architectures.
Ability to navigate ambiguity by breaking loosely defined or novel business problems into well-structured ML projects, defining metrics, setting up experiments, and iterating based on data.
Ability to collaborate effectively with cross-functional partners, tailoring complex technical concepts for non-technical stakeholders to drive alignment and decision-making across product, engineering, and go-to-market teams.
Ability to lead as the most senior technical IC on a team, independently owning high-impact technical initiatives, influencing architecture and roadmap decisions, and setting a high bar for code quality and model performance.
Ability to thrive in a fast-moving environment where combining rigorous experimentation with pragmatic engineering unlocks measurable value for advertisers, retailers, and end customers.

Responsibilities

Design, build, and productionize core ML models for CTR prediction, retrieval, conversion, and search/recommendation that directly drive performance for Moloco Commerce Media customers.
Lead end-to-end experimentation cycles, from problem framing and offline modeling to online A/B testing and iteration, to continuously improve auction, ranking, and recommendation quality.
Partner closely with product, engineering, and solutions teams to translate retail media and ecommerce requirements into robust ML solutions, features, and evaluation frameworks.
Drive technical excellence across the ML stack by improving pipelines, model serving, and monitoring for low-latency, high-throughput, and highly reliable inference at scale.
Mentor and uplevel other machine learning engineers through design reviews, knowledge sharing, and best practices in modern ML and MLOps.