AI Engineer

Pillsbury Winthrop Shaw Pittman•Nashville, TN

8d•Onsite

About The Position

The AI Engineer is responsible for designing, developing, testing, and deploying AI-enabled solutions that support the Firm’s legal and business operations. Working directly with attorneys, practice groups, and Firmwide Department teams, this role translates business needs and workflows into practical, scalable AI solutions using technologies such as Azure Databricks, Microsoft Azure, n8n, LangGraph, and Databricks Agent Bricks. As part of the Firm’s AI engineering team, the AI Engineer will contribute throughout the development lifecycle, from discovery and prototyping through production deployment, monitoring, and continuous improvement. The role will collaborate closely with Senior AI Engineers, Data Architects, DevOps teams, and other technical stakeholders while applying shared engineering standards and best practices. Engineers may bring complementary strengths in areas such as data engineering, application and front-end development, or applied data science, with assignments aligned to individual capabilities, development interests, and evolving business needs.

Requirements

Strong Python programming skills and demonstrated ability to develop maintainable, production-quality software using APIs, automated testing, version control, and code review.
Working knowledge of SQL and experience integrating application, data, and AI components.
Hands-on experience developing AI agents, agent-enabled applications, or multi-step AI workflows involving language models, enterprise information, APIs, or external tools.
Experience with prompt and context design, structured outputs, tool integration, state management, error handling, and human-in-the-loop workflows.
Practical experience with Azure Databricks and working knowledge of Microsoft Azure services used for application development, integration, or deployment.
Understanding of retrieval-augmented generation (RAG), embeddings, retrieval techniques, and the relationship between source quality and AI system performance.
Experience evaluating and troubleshooting AI applications, including analyzing execution traces, failures, task completion, quality, latency, or cost.
Understanding of production engineering practices, including application deployment, configuration management, secrets management, monitoring, automated testing, and troubleshooting.
Working knowledge of authentication, authorization, role-based access controls, and security considerations associated with enterprise AI applications and sensitive information.
Strong analytical and problem-solving skills with the ability to translate ambiguous business requirements into practical technical solutions.
Ability to communicate technical concepts and tradeoffs effectively to both technical and non-technical stakeholders.
Demonstrated collaboration, accountability, curiosity, and ability to work effectively with engineers, domain experts, and business users.

Nice To Haves

Experience deploying, monitoring, and supporting AI applications in a production environment.
Experience with one or more AI orchestration or workflow technologies, including n8n, LangGraph, Databricks Agent Bricks, or comparable platforms.
Experience with MLflow, LangSmith, or similar AI tracing, observability, and evaluation tools.
Advanced experience with Azure Databricks technologies, including Spark, Delta Lake, Unity Catalog, Lakebase, or Delta Sharing.
Experience designing and optimizing RAG solutions, including vector search, hybrid retrieval, filtering, reranking, document processing, metadata extraction, and retrieval evaluation.
Experience with graph databases, knowledge graphs, graph-based RAG, or techniques for identifying and linking entities and relationships across information sources.
Experience working with PostgreSQL or similar technologies to support persistent agent state, checkpoints, or memory.
Front-end or application development experience using TypeScript, JavaScript, React, or similar technologies.
Experience designing user-facing AI applications incorporating streaming responses, citations, progress indicators, feedback mechanisms, and human review.
Experience developing reusable AI tools and enterprise integrations, including Model Context Protocol (MCP) or comparable integration approaches.
Experience designing evaluation datasets, benchmarks, regression testing, automated scoring, or LLM-based evaluation methods.
Knowledge of statistical analysis or experimental design used to evaluate AI system performance and improvements.
Experience with Microsoft Entra ID, managed identities, and enterprise authentication and authorization models.
Experience working in legal services, professional services, technical consulting, or another environment involving sensitive information and complex business workflows.

Responsibilities

Partner with attorneys, practice groups, and business teams to identify AI opportunities, understand workflows, define requirements, and establish measurable success criteria.
Design, develop, test, and deploy AI agents and multi-step workflows using platforms and frameworks such as Azure Databricks, n8n, LangGraph, and Databricks Agent Bricks.
Integrate AI solutions with enterprise applications, APIs, governed data sources, and native Microsoft Azure services, incorporating appropriate human review and approval processes.
Develop prototypes in collaboration with users and translate successful concepts into reliable, scalable production applications.
Build and maintain evaluation datasets, automated tests, and quality measures to assess agent behavior, retrieval quality, tool-use accuracy, task completion, performance, latency, and cost.
Analyze execution traces, system behavior, and failure patterns to troubleshoot issues and continuously improve AI solutions.
Collaborate with Data Architecture and DevOps teams to support governed data access, application deployment, environment configuration, monitoring, and production support.
Design AI solutions with appropriate reliability safeguards, including error handling, retries, timeouts, stopping conditions, recovery processes, and controls designed to prevent unintended or duplicate actions.
Apply security, access controls, data protections, source traceability, and responsible AI practices throughout solution design and deployment, including protections against prompt injection, inappropriate access, and data leakage.
Engage with users following deployment to gather feedback, support adoption, measure business impact, and identify opportunities for continued enhancement.
Participate in code reviews, technical documentation, knowledge sharing, and the development of reusable components, standards, and engineering best practices.