Senior MLOps Engineer

Ignite•Suitland, MD

About The Position

Ignite IT Hub is seeking a senior MLOps Engineer to support reliable operation and controlled improvement of the U.S. Census Bureau's AWS-based LLM Autocoder. You will maintain software and data pipelines, automate testing and releases, and help resolve production issues. Working with AI/ML engineers and evaluators, you will make approved changes reproducible, observable, and recoverable across the American Community Survey (ACS) and related applications.

Requirements

Bachelor's degree in computer science, engineering, information technology, or a related field, or equivalent relevant professional experience.
8+ years in DevOps, cloud, software, or data engineering, including 3+ years operating ML systems or automated data-processing platforms.
Strong AWS and Linux experience, with Python or shell scripting and practical knowledge of cloud identity, access controls, storage, and monitoring.
Experience with Git, CI/CD, automated testing, environment configuration, dependency management, and reliable deployment and rollback practices.
Experience with ML artifact versioning or experiment tracking, production troubleshooting, and clear operational documentation; ability to coordinate across engineering and Government teams.

Nice To Haves

Experience operating LLM inference workflows, batch classification pipelines, or federal systems handling sensitive information.
Experience with infrastructure as code, containers, and managed AWS AI/ML services; relevant AWS certification.

Responsibilities

Maintain and improve software and data pipelines and continuous integration and delivery (CI/CD) workflows using Government-approved tools.
Version and trace code, model identifiers, prompts, configurations, data dependencies, and evaluation artifacts so results and releases can be reproduced.
Automate build, test, packaging, and deployment steps; integrate required validation evidence and approval checkpoints into release workflows.
Monitor processing health, failures, latency, throughput, and resource use; connect model-performance monitoring with metrics defined by the data science and evaluation team.
Diagnose production incidents and pipeline failures, document workarounds and root causes, and verify recovery with the AI/ML Engineer and Government staff.
Prepare release and rollback procedures, integrate Government-approved changes, and verify system behavior after deployment.
Apply access, secrets, logging, and sensitive-data handling controls; maintain runbooks, configuration records, and operational knowledge for cross-training and transition.