Artificial Intelligence Engineer

AI Engineer – Databricks Agents

Remote

Long-Term Contract

While professional experience and qualifications are key for this role, make sure to check you have the preferable soft skills before applying if required.

Brooksource is seeking an AI Engineer to support a high-visibility Nuclear Online Work Management initiative at one of our largest Fortune 200 energy clients. This program is focused on modernizing how nuclear operations teams plan, prioritize, and execute work by introducing Generative AI and Agentic AI capabilities into critical work management processes.

The AI Engineer will help build the first generation of production AI agents supporting this transformation, working hands-on within Databricks to develop intelligent solutions that can interpret operational changes, surface relevant information, recommend work prioritization, and help teams make faster, more informed decisions. These solutions will leverage both structured operational data and large volumes of unstructured procedures, policies, and technical documentation while maintaining the governance and human oversight required within a nuclear environment. This is a hands-on engineering role at the intersection of Generative AI, Agentic AI, RAG, Databricks, and nuclear operations. The ideal candidate has experience taking LLM-based applications from prototype through production and is excited about applying emerging AI technologies to complex, real-world operational challenges!

Requirements:

• 4+ years of experience building and supporting production software, data engineering, or AI/ML solutions within enterprise environments
• Strong hands-on development experience with Python and SQL, including building production-ready applications, data pipelines, APIs, or backend services
• 1+ year of hands-on experience developing applications utilizing Large Language Models (LLMs), Generative AI, or Agentic AI
• Demonstrated experience taking at least one LLM-based application from development through production deployment and supporting real-world users
• Experience designing and building Retrieval-Augmented Generation (RAG) solutions that leverage enterprise data, document repositories, and other governed information sources
• Strong hands-on experience with Databricks, including Unity Catalog, Delta tables, Databricks Jobs, serverless compute, and working within governed enterprise data environments
• Experience developing document extraction and processing pipelines that transform large volumes of unstructured documents into structured, searchable, and queryable data
• Understanding of AI agent architecture and design, including tool selection, permissions, grounding, orchestration, and defining appropriate boundaries for agent behavior
• Experience implementing AI guardrails and human-in-the-loop workflows to ensure recommendations are reviewed and approved before triggering downstream actions
• Experience evaluating AI and RAG solutions for accuracy, retrieval quality, latency, reliability, and cost, with the ability to use evaluation results to continuously improve production performance
• Understanding of AI observability, tracing, monitoring, and auditability within production environments, particularly when working with sensitive or highly governed enterprise data

Preferred Qualifications:

• Experience building Agentic AI or multi-agent workflows
• Experience with MLOps, AI tracing, monitoring, and evaluation
• Experience working with governed or sensitive enterprise data

Responsibilities

• Design, build, deploy, and support production AI agents within Databricks that improve nuclear work management and operational decision-making
• Develop RAG capabilities across structured enterprise data and unstructured procedures, policies, technical documentation, and operational records
• Build multi-agent workflows that interpret operational changes, identify impacted work, and recommend appropriate reprioritization or next steps
• Implement human-in-the-loop controls and guardrails to xhyhwjd ensure AI-generated recommendations are reviewed and approved before triggering downstream actions
• Build document extraction and processing pipelines that convert large volumes of unstructured records into governed, searchable, and queryable data
• Integrate AI agents with enterprise work management systems, operational data sources, APIs, and Databricks data products
• Develop testing and evaluation frameworks to measure retrieval quality, response accuracy, agent performance, and reliability before and after production deployment
• Implement AI monitoring, tracing, observability, and performance optimization to continuously improve accuracy, latency, reliability, and cost
• Identify opportunities where deterministic logic, SQL, or traditional reporting should be utilized instead of an LLM to improve reliability and efficiency
• Partner with engineering, data, architecture, nuclear operations, and business teams to translate operational use cases into scalable AI solutions
• Support AI governance, solution architecture, audit logging, disaster recovery, and technical documentation required for production deployment within a highly regulated nuclear environment
• Troubleshoot and continuously enhance production AI solutions as new use cases, data sources, and operational requirements are introduced