Staff Machine Learning Engineer

Mission

The Staff ML Engineer owns the design, deployment, and ongoing health of the machine learning systems that directly drive Kargo's revenue — from bid pricing and pacing to CTR prediction, viewability, and audience targeting. This role bridges data science and production engineering, turning models into reliable, scalable systems that perform under real-world ad tech conditions. When this role is firing, ML solutions ship faster, models stay healthy in production, and the business can trust the outputs driving auction decisions.

This is a remote role in Dublin, Ireland

Outcomes — What Success Looks Like in 6–12 Months

• Production ML models are reliable, monitored, and continuously improving — Active models have monitoring and alerting in place for drift and degradation; performance metrics are tracked and optimization is ongoing rather than reactive
• CI/CD pipelines accelerate model deployment — Model versioning, updates, and deployment are automated end-to-end; the time from model validation to production is measurably shorter than at hire
• At least one high-impact ML system is shipped and driving measurable business outcomes — A new or significantly improved model — bid optimization, CTR prediction, or audience targeting — is live in production and tied to a documented revenue or efficiency impact
• ML infrastructure is scalable and cost-efficient — Data pipelines, feature stores, and cloud tooling (AWS, Snowflake, Databricks) are optimized for both performance and cost; infrastructure decisions are made with operational sustainability in mind
• Cross-functional delivery is smooth and low-friction — Data science, engineering, and product teams are working from shared standards; integrations don't require heroic coordination and knowledge is documented rather than siloed

Skills — Core Technical Capabilities

Required

• 6+ years of experience building and deploying ML models in production environments — has owned the full lifecycle from training through inference, monitoring, and iteration
• Strong proficiency in Python and SQL for model development, data manipulation, and pipeline work; experience with Spark for large-scale distributed data processing
• Hands-on experience with AWS (S3, EC2, Lambda, SageMaker) and cloud-native ML workflows; comfortable provisioning and managing cloud infrastructure programmatically
• Familiarity with the MLOps stack — Databricks, Feature Stores, Kubernetes, Kubeflow, MLflow — and how these tools fit together in a production ML system
• Experience building both offline and online training and inference pipelines for real-time systems with latency and throughput constraints
• Strong understanding of CI/CD practices applied to ML — model versioning, automated testing, deployment pipelines, and rollback strategies

Preferred

• Ad tech or digital advertising experience — familiarity with auction dynamics, bid optimization, CTR/viewability prediction, or audience targeting models
• Experience with Go for performance-sensitive production services

Competencies — Behaviors We Like to See

Production-First Mindset

• Builds models with deployment, monitoring, and maintenance in mind from the start — not as an afterthought once the science is done
• Sets up alerting and drift detection proactively; treats model health as an ongoing operational responsibility, not a launch checklist item

Rigorous Technical Execution

• Brings discipline to code quality, version control, and testing in ML workflows — raises the bar on engineering standards across the team through code reviews and shared practices
• Optimizes for the right tradeoffs — accuracy, latency, cost, reliability — and makes those tradeoffs explicitly rather than defaulting to one dimension

Cross-Functional Translation

• Communicates complex ML concepts clearly to data scientists, e