Why This Job is Featured on The SaaS Jobs
This Machine Learning Engineer role stands out in SaaS because it sits at the point where product value is delivered through real-time inference, not offline experimentation. The remit spans streaming telemetry, multi-tenant model customization, and low-latency serving, which are recurring challenges in data-intensive SaaS platforms that must make decisions continuously across a broad customer base.
From a career perspective, the work builds durable SaaS skills in production ML: designing pipelines that can be retrained and redeployed safely, reasoning about failure modes in distributed systems, and optimizing end-user latency as a first-class constraint. Experience translating evolving fraud patterns into features and models also develops a pragmatic approach to applied ML, where model quality is measured against operational outcomes and shifting adversarial behavior.
The role is best suited to engineers who prefer owning end-to-end ML systems and are comfortable moving between backend performance work and modeling choices. It will fit professionals who enjoy cross-functional problem framing with infrastructure, data science, and product partners, and who want their ML work to be tightly coupled to production reliability and measurable platform impact.
The section above is editorial commentary from The SaaS Jobs, provided to help SaaS professionals understand the role in a broader industry context.
Job Description
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
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 and follow us on LinkedIn.
