Lead MLOps Engineer

EPAM Systems; Colombia; Remote; COP 287,954,000 - 415,933,000; Full time

Benefits offered by this job

Healthcare benefits

Paid time off

Learning opportunities

Job summary

EPAM Systems, Inc. in Colombia seeks a Lead MLOps Engineer to head an MVP engagement for a major AAA game publisher. You will design a reproducible platform with full lineage from day one and ensure production readiness across ML components.

You will mentor engineers, define standards for model lifecycle, provenance, and calibration harness, and drive cross‑franchise collaboration to deliver scalable, auditable ML systems.

Qualifications

• 5+ years in MLOps or ML platform engineering.
• Deep expertise in ML model lifecycle management, including versioning and rollback.
• Strong background in architecting signal computation pipelines.
• Advanced proficiency in calibration methodology and holdout discipline.
• Advanced skills in Python and SQL for ML pipeline automation.

Responsibilities

• Define the MLOps architecture and strategy for the platform across franchises.
• Own signal catalogue operations end-to-end and define provenance standards.
• Set strategy for Semantic Vector Index versioning and controlled reindex paths.
• Design and own the Back-test & Calibration Harness with as-of temporal filtering.
• Lead model generation, experimentation framework, and cross-franchise governance.
• Mentor engineers and review designs, representing MLOps in cross-team discussions.

Skills

MLOps leadership

Python

SQL

Model monitoring

Drift detection

Experimentation framework

Cross-functional collaboration

Mentorship

Versioning

Back-test harness

Tools

Snowflake Snowpark

pgvector

Job description

We are seeking a Lead MLOps Engineer to head an MVP engagement with a major AAA game publisher, building a test intelligence platform for two game franchises in parallel. A core design principle is full re-derivability and model lineage from day one — every run must be replayable from its stored configuration version and feed read positions. The signal catalog feeds a scoring strategy engine with versioned configurations, and calibration sweeps over historical data produce suggested weight updates surfaced directly in the Settings View.This role sets the technical direction for the ML and signal components, ensuring they are production-ready, reproducible, and improvable over time, forming the foundation of the system's long-term value as franchise history accumulates and models are refined.The Lead MLOps Engineer will define standards, mentor engineers, and act as the primary technical authority for all MLOps practices across both franchise workstreams.ResponsibilitiesDefine the overall MLOps architecture and strategy for the platform, establishing standards for reproducibility, lineage, and model lifecycle management across both franchisesOwn Signal Catalogue operations end-to-end: architect signal refresh orchestration triggered by feed read-position advances, define grain translation policies between per-test, per-area, and per-run signal families, and establish provenance capture standards across all 8 signalsSet the strategy for Semantic Vector Index versioning: partner with the Lead AI Developer on model and dimension stamp conventions; design and govern the controlled reindex path when the enterprise AI gateway model changesArchitect, deliver, and own the Back-test & Calibration Harness: as-of temporal filtering across all record families, replay runner, look-ahead spot audit, and configuration sweep runnerEstablish and enforce holdout patch-set discipline, define configuration sweep methodology over route limits, thresholds, weights, and Composition setting; oversee catch-rate vs. scope analysis per candidate configuration and approve winning configurations as suggested-weight proposals into the Settings ViewLead Model Generation & Experimentation strategy: define the systematic experimentation framework over scoring strategy configurations, guiding the team on which signal weights and route combinations yield the best catch-rate vs. scope trade-offGovern model lineage across configuration versions for both franchises, ensuring cross-franchise consistency and auditabilityDesign and oversee the MLOps Monitor and Data-Health Monitor: catch-rate floor monitoring, run-behaviour drift counters (per-run candidate volumes per route, score distribution vs. usual range), data-health telemetry across all ingestion channelsDefine exploration cadence policy: unbiased random-sample injection with provenance ensuring exploration entries are never counted as model recommendationsAuthor and own operator runbook sections covering signal refresh, calibration campaigns, model generation runs, and embedding reindex proceduresMentor engineers on MLOps best practices, review technical designs, and represent the MLOps function in cross-team architecture discussions with data engineering, AI development, and product stakeholdersRequirements5+ years of experience in MLOps or ML platform engineering, with a proven track record of leading technical initiatives end-to-endDeep expertise in ML model lifecycle management, including versioning, configuration management, and rollback, with experience defining organization-wide standardsStrong background in architecting signal computation pipelines, covering scheduled refresh, provenance capture, and grain translationAdvanced proficiency in calibration methodology: holdout discipline, configuration sweep design, and catch-rate vs. scope measurementExpert knowledge of as-of temporal data systems or back-test harness design and operationAdvanced skills in Python and SQL for ML pipeline automation, with experience setting coding and design standards for a teamDeep competency in model monitoring, including drift detection, catch-rate floor monitoring, and run-behavior drift countersProven ability to lead cross-functional collaboration with data engineers and AI developers on feature alignment and shared roadmapsStrong track record of authoring calibration procedures, signal definitions, and operational runbooks that scale across teamsSolid experience with Spec Driven Development, ideally as a methodology advocate or championDemonstrated mentorship and technical leadership experience, including code review, design review, and guiding mid-to-senior engineersExcellent written and verbal communication skills in English (B2+ level)Nice to haveFamiliarity with Snowflake ML or Snowpark in production settingsProven showcase of embedding model versioning and controlled reindex orchestration at scaleOperational leadership with pgvector or vector store managementBackground in gaming domain or QA toolingWe offerInternational projects with top brandsWork with global teams of highly skilled, diverse peersHealthcare benefitsEmployee financial programsPaid time off and sick leaveUpskilling, reskilling and certification coursesUnlimited access to the LinkedIn Learning library and 22,000+ coursesGlobal career opportunitiesVolunteer and community involvement opportunitiesEPAM Employee GroupsAward-winning culture recognized by Glassdoor, Newsweek and LinkedInEPAM is an Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, age, sexual orientation, gender identity or expression, disability, protected veteran status, or any other characteristic protected by applicable law.