Staff Data Scientist, Applied ML — Real-Time ML & RAG

Jobber; Canada; Hybrid; CAD 146,000 - 197,000; Full time

Job summary

Jobber is seeking a Staff Data Scientist, Applied Machine Learning to own end-to-end ML systems, including production pipelines, monitoring, and retraining, applied to large-scale SaaS event data. You will design and evaluate RAG systems on a graph backbone, partner with leadership, and drive ML engineering standards across the team.

You will work closely with product, analytics, and BI teams to translate complex problems into scalable ML solutions that impact real business outcomes at scale.

Qualifications

• Production ML end-to-end experience in real systems.
• Strong foundation in statistics and model evaluation.
• Expert SQL and production-grade Python skills.
• Hands-on experience deploying ML models in production with monitoring, drift detection, and retraining.
• Experience with large-scale data in production and modern ML platforms.

Responsibilities

• Design, build, and maintain ML models and pipelines for production use.
• Develop and evaluate RAG systems on graph data to ensure retrieval quality.
• Own end-to-end ML lifecycle including deployment, monitoring, and retraining.
• Establish regression testing and evaluation standards for ML/LLM projects.
• Collaborate with senior leadership and cross-functional teams to align data science with business goals.

Skills

Production ML

Statistics

SQL

Python

LLMs

RAG

Stakeholder alignment

Tools

Snowflake

Airflow

Docker

Kubernetes

REST/gRPC

Job description

Jobber is seeking a Staff Data Scientist, Applied Machine Learning to own end-to-end ML systems, including production pipelines, monitoring, and retraining, applied to large-scale SaaS event data. You will design and evaluate RAG systems on a graph backbone, partner with leadership, and drive ML engineering standards across the team.

You will work closely with product, analytics, and BI teams to translate complex problems into scalable ML solutions that impact real business outcomes at scale.