Machine Learning Engineer

THE ROLE:

As a Machine Learning Engineer at Sift, you will bridge the gap between data science and large-scale distributed systems. You won’t just train models in isolation; you will build end-to-end pipelines that extract signals, train custom models per merchant, and serve predictions at production scale with low latency. You will work on an automated machine learning ecosystem that dynamically recalibrates models based on streaming global telemetry data.

WHAT YOU'LL DO:

- Model Development & Refinement: Design, build, and deploy online machine learning models (including ensemble methods, deep learning, transformer architectures and graph-based models) to catch evolving fraud vectors in real time.

- Feature Engineering at Scale: Engineer high-frequency time-series features from over 1 trillion behavioral events, optimizing for low-latency signal extraction and pattern recognition.

- Production MLOps: Maintain and enhance our automated model training and deployment infrastructure, ensuring frictionless continuous integration and continuous deployment (CI/CD) of newly trained models.

- System Optimization: Write high-performance code to minimize scoring latency at runtime, ensuring our core ML services scale seamlessly across distributed databases.

- Collaborative Innovation: Work cross-functionally with Core Infrastructure, Product Management, and Data Science teams to translate business-level fraud patterns into robust algorithmic solutions.

WHAT WE ARE LOOKING FOR (REQUIREMENTS):

- Experience: 4+ years of professional experience building and deploying large-scale machine learning models into high-traffic production environments.

- Solid Programming Foundations: Strong proficiency in Java or Scala (for our production backend) as well as Python (for data analysis and model prototyping).

- Distributed Systems & Big Data: Practical experience with Databricks and big data processing frameworks like Apache Spark, Apache Flink, or Hadoop, and working with NoSQL data stores like Bigtable.

- Strong Mathematical Foundations: Deep understanding of statistical modeling, probability, and standard machine learning algorithms (e.g., XGBoost, Random Forests, Neural Networks, and Clustering techniques).

- System Design Mentality: Ability to reason through data consistency, pipeline failures, and performance constraints in a distributed, multi-tenant cloud environment (GCP).

BONUS POINTS (PREFERRED QUALIFICATIONS):

- Experience explicitly in the fraud detection, risk mitigation, or cyber-security domains.

- Deep knowledge of streaming architectures (e.g., Apache Kafka).

- Familiarity with containerization and orchestration tools like Docker and Kubernetes.

- Familiarity with leveraging AI coding assistants (e.g., Claude Code) to accelerate development and model prototyping

Please note: final stage candidates may be asked to travel for in-person final round interviews.

Let’s build it together:

At Sift, we are intentionally building a diverse, equitable, and inclusive workplace. We believe that diversity drives innovation, equity is a fundamental right, and inclusion is a basic human need. We envision a place where all Sifties feel secure sharing their authentic selves and diverse experiences with their teams, their customers, and their community – ultimately using this empowerment and authenticity to build trust and create a safer Internet.

This document provides transparency around how Sift handles the personal data of job applicants: https://sift.com/recruitment-privacy

A little about us: Sift is the AI-powered fraud platform securing digital trust for leading global businesses. Our deep investments in machine learning and user identity, a data network scoring 1 trillion events per year, and a commitment to long-term customer success empower more than 700 customers to grow fearlessly. Global brands rely on Sift to unlock growth and deliver seamless consumer experiences. Visit us at sift.com http://sift.com and follow us on LinkedIn https://www.globenewswire.com/Tracker?data=XHeK0v8NcNrEkwcDe8QxwpZeCkdQqNyKlni83U-CUmrprdKXWpVlYOAbVzwe2OmlwIUN-q4HXk4hf_dazpHx2NMM1CW_SYj740q9mxXNQI4=.