MLOps Engineer — AI/ML Systems Deployment (TS/SCI Preferred)

Rackner is hiring an MLOps Engineer to move AI/ML systems from prototype → deployment → operational use in a secure, mission-focused environment.

This is not a research role—this is where models become reliable, repeatable, auditable systems that run in real-world conditions.

This role is ideal for engineers who want to:

• Work across AI/ML, Kubernetes, infrastructure, and mission systems
• Own deployed systems, not just experiments
• Build high-demand MLOps expertise in secure and constrained environments
• Deliver technology that is used, trusted, and operational

You will help operationalize AI/ML capabilities where reliability, performance, and trust matter most.

• Active TS/SCI clearance
• Active Secret clearance with eligibility for upgrade
• Familiarity with ML lifecycle tools such as MLflow, Kubeflow, Airflow, Argo, ClearML, or similar
• Background in model serving, inference APIs, or deploying ML systems in production
• Exposure to LLMs, transformer-based models, computer vision, NLP, or applied AI solutions
• Hands-on work with Kubernetes-based ML workloads
• Knowledge of observability and monitoring tools such as Prometheus, Grafana, or OpenTelemetry
• Experience in DoD, defense, intelligence, regulated, or mission-critical settings
• Work in edge, offline, air-gapped, low-bandwidth, D-DIL, or limited-compute environments

• Active TS/SCI clearance strongly preferred
• Candidates with an active Secret clearance may be considered and supported for upgrade
• Candidates without an active clearance must be:
  
  • U.S. citizens
  • eligible to obtain and maintain a clearance
  • able to work in a CAC-enabled or secure environment

Rackner is a software consultancy that builds cloud-native solutions for startups, enterprises, and the public sector. We are an energetic, growing team focused on solving complex problems through:

• Distributed systems
• DevSecOps
• AI/ML
• Cloud-native architecture

Our approach is cloud-first, cost-effective, and outcome-driven, delivering systems that scale and perform in real-world environments.

• Deploy AI/ML models and ML-enabled applications into secure, real-world environments
• Move workflows from experimentation into containerized, repeatable deployment pipelines
• Support batch and real-time inference architectures
• Bridge model development, software engineering, and platform operations

• Build and operate production-grade ML pipelines
• Support model versioning, lineage, reproducibility, and lifecycle governance
• Work with tools such as MLflow, Kubeflow, Airflow, Argo, ClearML, or similar platforms

• Deploy and support Kubernetes-based ML workloads
• Containerize models, pipelines, and services using Docker or similar tools
• Support CI/CD, automation, and repeatable deployment patterns for AI/ML systems

• Monitor model and system performance after deployment
• Support observability using tools such as Prometheus, Grafana, OpenTelemetry, or similar
• Detect and resolve issues related to latency, reliability, drift, degradation, or resource usage

• Help deploy AI/ML systems in secure, CAC-enabled, or constrained environments
• Support limited compute, restricted data, degraded connectivity, and other operational constraints
• Optimize systems for reliability and usability beyond ideal lab conditions

• Develop runbooks, deployment documentation, and operational playbooks
• Build systems that can be understood, maintained, and operated by others

• U.S. citizenship
• Background in deploying ML systems, AI-enabled applications, or production software
• Strong programming skills in Python
• Hands-on work with Docker, containers, or containerized deployment
• Familiarity with Kubernetes or cloud-native environments
• Understanding of CI/CD, automation, or pipeline-based delivery
• Clear communication of technical decisions, tradeoffs, and ownership
• Ability to operate in a CAC-enabled or secure environment

If you are an engineer who wants to move from building models or platforms to owning deployed AI/ML systems, we would like to connect.