Senior MLOps Engineer — Reproducible ML for Games

EPAM Systems; Chile; On-site; CLP 42,000,000 - 60,000,000; Full time

Benefits offered by this job

Healthcare benefits

Paid time off & sick leave

Upskilling & certification courses

Global career opportunities

Volunteer & community involvement

Job summary

EPAM Systems is seeking a Senior MLOps Engineer to join an MVP engagement for a high-profile game publisher, building a test intelligence platform for two game franchises. The role emphasizes full re-derivability, model lineage, and reproducibility from day one as the system evolves with franchise history.

You will own signal catalog operations, manage semantic vector index versioning, and design back-test/harness workflows while enforcing rigorous calibration and holdout discipline to drive

Qualifications

• 3+ years in MLOps or ML platform engineering.
• Expertise in ML model lifecycle management including versioning/configuration/rollback.
• Background in signal computation pipeline design, coverage of refresh/provenance.
• Proficiency in calibration methodology: holdout discipline, sweep design, catch-rate vs. scope.
• Knowledge of as-of temporal data systems or back-test harnesses.
• Skills in Python and SQL for ML pipeline automation.
• Competency in model monitoring including drift detection and run-behavior drift counters.
• Ability to collaborate with data engineers and AI developers on feature alignment.
• Document calibration procedures, signal definitions, and operational runbooks.

Responsibilities

• Own Signal Catalogue operations and provenance capture across signals.
• Coordinate with AI developers on model/dimension stamp conventions and reindex paths.
• Design, implement, and own the Back-test & Calibration Harness.
• Enforce holdout patch-set discipline and publish winning configurations as weight proposals.
• Lead Model Generation & Experimentation with systematic experimentation framework.
• Maintain model lineage across configuration versions for both franchises.
• Implement MLOps Monitor and Data-Health Monitor across all ingestion channels.
• Support exploration cadence with unbiased random-sample injections.
• Contribute to operator runbooks covering signal refresh, calibration campaigns, and embedding reindex procedures.

Skills

MLOps

ML lifecycle

Python

SQL

Model monitoring

Collaboration

Documentation

Spec-Driven Dev

English B2

Tools

MLflow

Kubeflow

Snowflake ML

Snowpark

pgvector

Job description

EPAM Systems is seeking a Senior MLOps Engineer to join an MVP engagement for a high-profile game publisher, building a test intelligence platform for two game franchises. The role emphasizes full re-derivability, model lineage, and reproducibility from day one as the system evolves with franchise history.

You will own signal catalog operations, manage semantic vector index versioning, and design back-test/harness workflows while enforcing rigorous calibration and holdout discipline to drive