Principal AI Engineer
MARS Solutions Group · United States · 5 days ago
RemoteRemoteEngineeringFull-time
About the role
MARS Solutions Group is seeking a Principal AI Engineer for a 3-4 month engagement with our client, a Building Automation industry leader. The role involves embedding within the Controls Software team to develop foundational architecture assets and deployment playbooks for AI adoption across their smart building platform.
Responsibilities
- Collaborate daily with Controls SW engineers, architects, and product managers
- Partner with internal domain experts to onboard to the platform
- Drive the architecture, validate assumptions with the team, and produce deliverables iteratively
- Participate in team rituals (standups, design reviews, retrospectives)
Requirements
- AI Systems & Architecture: 10+ years of hands-on software or systems engineering experience, with at least 6 years focused on AI/ML in production environments
- Proven experience designing and deploying AI/ML systems at scale from data ingestion through inference and monitoring
- Deep knowledge of MLOps: model deployment pipelines, versioning, observability, drift detection, and continuous improvement
- Experience with edge-to-cloud AI execution strategies: balancing latency, cost, and resiliency across distributed environments, including LLM cost optimization (model selection, caching, routing)
- Strong command of data pipeline architecture, time-series data, event-driven systems, and API/microservices patterns
Preferred Qualifications
- Experience in industrial, OT, or IoT environments (building automation, manufacturing, energy, or similar)
- Familiarity with protocols such as BACnet, MQTT, Modbus, or OPC UA
- Exposure to cybersecurity frameworks in OT environments (e.g., IEC 62443, NIST CSF)
- Experience with AI use cases in buildings or critical infrastructure: FDD, energy optimization, predictive maintenance, alarm intelligence
- Experience with containerization, CI/CD tooling, and observability platforms
- Experience operationalizing AI safety guardrails, content filtering, and governance controls in production