Machine Learning Engineer

Zume Search Llc; United States; On-site; USD 150,000 - 190,000; Full time

Job summary

Zume Search Llc is seeking an experienced Machine Learning Engineer to own end-to-end ML model development for digital trust and security. You will build, deploy, and monitor production models on massive datasets from online interactions, shaping detection capabilities and threat insights.

Ideal candidates have 5+ years in production ML, strong Python and ML framework skills, and experience with security-related data and cloud environments.

Qualifications

• 5+ years of experience building and deploying machine learning models in production.
• Strong Python skills and experience with PyTorch, TensorFlow, Scikit-Learn, XGBoost, LightGBM, or similar frameworks.
• Demonstrated success improving model performance on noisy, real-world datasets.
• Experience with SQL, cloud data warehouses, and AWS environments.
• Knowledge of anomaly detection, clustering, or other unsupervised learning techniques.
• Strong analytical skills and the ability to move from raw data to actionable insights.
• Experience in cybersecurity, fraud detection, trust & safety, threat intelligence, ad tech, or related domains is highly desirable.

Responsibilities

• Build and maintain ML models to detect malicious activity and emerging online threats.
• Develop and improve classification systems, training datasets, and model evaluation frameworks.
• Monitor model performance, drift, FP, and FN in production environments.
• Conduct research and data analysis to uncover new signals and attack patterns.
• Partner with product and engineering to turn findings into customer-facing capabilities.
• Create reporting and insights to help stakeholders understand threat activity and detection coverage.
• Contribute to research and thought leadership around emerging digital risks.

Skills

Machine learning

Python

PyTorch

TensorFlow

Scikit-Learn

XGBoost

LightGBM

SQL

AWS

Anomaly detection

Clustering

Tools

PyTorch

TensorFlow

Scikit-Learn

XGBoost

LightGBM

Job description

I'm partnering with a high-growth technology company that's using machine learning, large-scale data analysis, and advanced detection techniques to solve complex digital trust and security challenges.

This is a unique opportunity to join a small, highly collaborative team where you'll have significant ownership and the ability to influence both product direction and technical strategy. If you enjoy finding signals in noisy data, building production-grade ML systems, and solving problems that don't have obvious answers, this could be a great fit.

What Makes This Opportunity Different?

This isn't a role where you'll simply train models and hand them off. You'll own the entire lifecycle, from identifying valuable signals and developing models to deploying them into production and measuring their effectiveness against real-world threats.

You'll work with massive datasets generated from millions of online interactions and have the opportunity to uncover patterns and risks that others miss.

What You'll Be Doing

• Build and maintain machine learning models used to detect malicious activity and emerging online threats.
• Develop and improve classification systems, training datasets, and model evaluation frameworks.
• Monitor model performance, drift, false positives, and false negatives in production environments.
• Conduct research and data analysis to uncover new signals, trends, and attack patterns.
• Partner with product and engineering teams to turn findings into customer-facing capabilities.
• Create reporting and insights that help stakeholders understand threat activity and detection coverage.
• Contribute to research and thought leadership around emerging digital risks.

What We're Looking For

• 5+ years of experience building and deploying machine learning models in production.
• Strong Python skills and experience with PyTorch, TensorFlow, Scikit-Learn, XGBoost, LightGBM, or similar frameworks.
• Demonstrated success improving model performance on noisy, real-world datasets.
• Experience with SQL, cloud data warehouses, and AWS environments.
• Knowledge of anomaly detection, clustering, or other unsupervised learning techniques.
• Strong analytical skills and the ability to move from raw data to actionable insights.
• Experience in cybersecurity, fraud detection, trust & safety, threat intelligence, ad tech, or related domains is highly desirable.

The Ideal Candidate

You're naturally curious and ask, "What don't we know yet?" before reaching for a model. You enjoy digging into messy data, identifying meaningful patterns, and translating findings into practical solutions. You're comfortable operating independently, communicating with both technical and non-technical stakeholders, and making a meaningful impact on a small team.