Senior MLOps Engineer

Senior Mlops Engineer

• Department: Engineering
• Employment Type: Flexible
• Location: Costa Rica

Description

Newfire Global Partners is a leading technology firm that specializes in building transformative software solutions for some of the world’s most innovative companies. With a presence across four continents, Newfire Global brings deep expertise in digital healthcare, AI-driven analytics, and enterprise technology. The firm’s track record of delivering scalable, high-impact solutions has made it a trusted partner for organizations seeking to drive meaningful change through technology.

We are passionate about the purpose-driven mission to help improve the quality of care for patients and are building a collaborative, innovative, and inclusive culture. We are a fully funded company founded by serial entrepreneurs with a stable client base.
Opportunity for impact

Newfire Global Partners, a leader in developing disruptive healthcare technology, collaborates with Fortune 500 companies and start-ups to drive transformation. 
Position Overview The Senior MLOps Engineer is an IC role that designs, automates, and operates the end-to-end ML/LLM production lifecycle by promoting and implementing MLOps practices. You will design cloud native infrastructure and build the CI/CD and IaC backbone for data, model, and inference workflows; build reusable testing and evaluation frameworks, harden runtime environments; implement safe release/rollback; and drive observability and cost efficiency at scale on AWS and Databricks.

Your Day-To-Day Activities:

Essential DutiesInclude, but are not limited to, the following:

• Own productionizing models—from tracked experiments to governed releases—ensuring resilient services with clear SLOs, runbooks, and fast, safe rollbacks.
• Build automation-first delivery: reproducible builds, layered tests, and environment promotion via GitLab CI and Terraform-based IaC.
• Engineer scalable serving: batch and real-time inference on EKS/ECS/Lambda and Databricks Model Serving with probes, autoscaling, and canary/blue-green deployments.
• Instrument end-to-end observability (data, model, system); detect drift/regressions; lead incidents and post-mortems that drive durable fixes.
• Partner across teams to translate requirements into designs, ADRs, and change plans; balance security, privacy, cost, and performance tradeoffs.
• Continuously reduce toil through automation, optimize model/GPU/LLM cost, and evolve templates/playbooks for repeatable delivery.

Please note that employment will be contingent upon providing documentation verifying your legal work authorization in the country of residence, in accordance with applicable law.

You’Re A Perfect Match If You Have:

Minimum Qualifications:

• Bachelor’s degree in Computer Science, Engineering, Data Science, or a related field and 3+ years of relevant experience as outlined in the essential duties; or High School Diploma/General Education Degree and 6+ years of relevant experience as outlined in the essential duties in lieu of Bachelor’s Degree.
• 3+ years operating ML systems in production (MLOps).
• Experience with Python for ML engineering (packaging, typing, testing, performance)
• Experience developing GitLab CI for ML/GenAI (multi-stage pipelines, artifacts, evaluation/security gates) and Terraform for ML/GenAI (reusable modules, drift detection); secure packaging & containerization.    
• Experience deploying and operating compute for ML (EKS/ECS/Lambda), and secure data access patterns (S3/VPC/IAM/KMS, private endpoints) 
• Experience implementing MLflow tracking, model registry & governed promotion, packaging & deployment to multi-target runtimes.  
• Experience operating real-time + batch/streaming inference workloads, ML observability, layered testing (unit/integration), workflow orchestration, and cost optimization.  
• Experience designing and implementing IAM least-privilege, secrets/key management for CI/CD pipelines; privacy and compliance awareness. 

Preferred Qualifications:

• Advanced GitLab CI (dynamic child pipelines, components, cross-project triggers, security scans, compliance gates).  
• Advanced Terraform (policy-as-code, gated plan/apply, environment promotion).  
• Advanced real-time serving (multi-tenant routing, dynamic model loading) and SLO-driven rollback/automation.  
• Databricks governance (Unity Catalog, lineage) and feature platform approval/reuse workflows.