MLOps Engineer

Evlo AI; Chicago (IL); On-site; USD 120,000 - 180,000; Full time

Job summary

Evlo AI is seeking an experienced MLOps/ML engineering professional to own end-to-end ML pipelines, training, and deployment. You will build scalable serving, implement robust CI/CD, and oversee observability for production models.

You will collaborate with data scientists to translate experimental code into robust systems, optimize GPU and cloud resources, and ensure reliability with SLOs and postmortems. This role is based in Chicago for on-site operations.

Qualifications

• 3–6 years of experience in MLOps, ML engineering, or backend/platform engineering with production ML systems.
• Strong Python skills plus hands-on experience with containerization and orchestration (Docker, Kubernetes).
• Proficiency with at least one ML pipeline/workflow tool: Kubeflow, MLflow, Airflow, or SageMaker Pipelines.

Responsibilities

• Design and maintain end-to-end ML pipelines for training, validation, and deployment.
• Build and scale model serving infrastructure with Docker, Kubernetes, and frameworks like KServe, Seldon, or Triton.
• Implement CI/CD for ML systems — automated testing, model versioning, canary deployments, and rollback strategies.
• Establish monitoring and observability for production models: data drift detection, latency/throughput dashboards, and automated alerting with Prometheus and Grafana.
• Manage GPU compute and cloud resources across AWS, GCP, or Azure, optimizing cost and utilization for training and inference workloads.
• Collaborate with data scientists to productionize models, translating experimental code into robust, tested engineering systems.
• Drive platform reliability standards: SLOs, incident response, and postmortems for ML serving infrastructure.

Skills

MLOps

Python

Docker/Kubernetes

CI/CD for ML

Cloud platforms

Monitoring

Education

BSc in CS/Engineering

Tools

Docker

Kubernetes

Kubeflow

MLflow

Airflow

SageMaker Pipelines

Terraform

Prometheus

Grafana

Feast

Tecton

Kafka

Flink

vLLM

TensorRT-LLM

Triton

Job description

About The Role

The role owns the infrastructure that makes ML work in production: training pipelines, model serving, monitoring, and the CI/CD systems that take models from notebook to reliable, scalable deployment.

The team partners closely with ML engineers and data scientists to eliminate friction between experimentation and production — building the automated pipelines, observability, and infrastructure standards that let models ship fast and stay reliable.

Key Responsibilities

• Design and maintain end-to-end ML pipelines for training, validation, and deployment using tools like Kubeflow, MLflow, or Airflow
• Build and scale model serving infrastructure with Docker, Kubernetes, and frameworks like KServe, Seldon, or Triton, supporting both batch and low-latency inference
• Implement CI/CD for ML systems — automated testing, model versioning, canary deployments, and rollback strategies
• Establish monitoring and observability for production models: data drift detection, latency/throughput dashboards, and automated alerting with Prometheus and Grafana
• Manage GPU compute and cloud resources across AWS, GCP, or Azure, optimizing cost and utilization for training and inference workloads
• Collaborate with data scientists to productionize models, translating experimental code into robust, tested engineering systems
• Drive platform reliability standards: SLOs, incident response, and postmortems for ML serving infrastructure

What We Are Looking For

• 3–6 years of experience in MLOps, ML engineering, or backend/platform engineering with production ML systems
• Strong Python skills plus hands-on experience with containerization and orchestration (Docker, Kubernetes)
• Experience with at least one ML pipeline/workflow tool: Kubeflow, MLflow, Airflow, or SageMaker Pipelines
• Proficiency with cloud infrastructure (AWS, GCP, or Azure), including GPU instance management and IaC tools like Terraform
• Practical experience with model serving and monitoring in production, including drift detection and performance tuning
• Bachelor's degree in Computer Science, Engineering, or equivalent practical experience
• Bonus: Experience with LLM inference infrastructure (vLLM, TensorRT-LLM, Triton), feature stores (Feast, Tecton), or streaming data systems (Kafka, Flink)