Data Engineer​/ML Engineer

This role requires a minimum of four (4) days per week working onsite at EnStream’s head office in Toronto; this requirement may be changed at management’s discretion.

EnStream is a leader in secure digital identity and mobile data intelligence, working to advance the future of digital trust in Canada. We build innovative data‑driven models that enhance the integrity, reliability, and safety of digital identity ecosystems. Our latest initiative leverages advanced data science
, machine learning
, and deep learning to further grow and sustain digital trust across Canada.

Our mission is to empower frictionless trust in every interaction. EnStream is dedicated to increasing trust and convenience for Canadians using real‑life, verified identities and network data held by trusted telco networks. At EnStream, every team member plays a critical role in shaping our strategy and delivering meaningful impact across industries.

We’re accelerating two high‑priority fraud detection and prevention initiatives and need a hands‑on Data & ML Engineer for 3 months (with potential extension). You’ll focus on building and hardening data pipelines, feature‑ready datasets, and reproducible unsupervised ML workflows to surface anomalies and behavioral clusters across baseline and partner data.

• Ingest and integrate internal and partner data into governed AWS storage/compute layers, including schema mapping, normalization, and automated validation checks
• Build production‑grade ETL/ELT pipelines (cleaning, joins, deduplication, entity resolution as needed) that generate feature tables and curated datasets for downstream modeling
• Develop reproducible unsupervised ML pipelines (Isolation Forest, K‑Means, DBSCAN) with modular components, configuration‑driven parameters, and end‑to‑end traceability
• Operationalize outputs into consumable data products (cluster/anomaly tables, feature outputs, dashboards‑ready datasets) with reliable refresh patterns
• Deliver engineering‑ready artifacts (notebooks, pipeline specs, data dictionaries, model configs, runbooks) to enable handoff and continued iteration
• Ensure ML engineering hygiene: versioned code/data, reproducible environments, experiment tracking, and documentation of transformations and model runs

• Strong Python (PySpark and/or pandas) and SQL
  , with a proven track record of building reliable, scalable data transformations and feature pipelines
• Hands‑on data engineering and ML engineering on AWS (e.g., S3, Glue/Athena/EMR, Redshift, Sage Maker), including orchestration, monitoring, and cost/performance optimization
• Experience building reproducible ML pipelines for unsupervised learning (clustering/anomaly detection), including feature generation, configuration management, and experiment/run tracking
• Ability to execute in a fast‑paced, time‑boxed environment—shipping high‑impact pipelines and ML workflows quickly while maintaining engineering quality and maintainability

• Prior data science experience (or strong applied analytics background) to help frame problems, validate assumptions, and interpret model outputs
• Solid understanding of the end‑to‑end AI/ML lifecycle
  , including problem definition, data readiness, feature engineering, training/experimentation, evaluation, deployment considerations, and monitoring/drift
• Familiarity with MLOps practices and tooling (e.g., experiment tracking, model registries, CI/CD for ML, reproducible environments) to support scalable iteration and handoff

• Contribute to a national‑scale initiative defining the future of digital trust in Canada
• Work on cutting‑edge fraud detection applications using real‑world identity data
• Collaborate with a highly skilled, cross‑functional team