Applied AI Field Engineer

Department: IT Location: Salt Lake City, UT Description Trilon is building a supercharged, technology-enabled future for our people and partners. The Applied AI Engineer plays a critical role in that mission by building the AI-powered features that enable our tools to compress real engineering labor across our operating companies. This role sits at the intersection of software engineering and applied AI, focused on designing and implementing the intelligence layer of our products. You translate product requirements and architectural patterns into working AI capabilities by building prompt frameworks, retrieval-augmented generation pipelines, and agent-based workflows that operate against real engineering data and deliverables. Working within a product pod, you partner closely with the Lead Engineer, Software Engineer, and QA Engineer to deliver production-ready solutions. You own how the system reasons, including prompt design, context management, model integration, and orchestration logic. You also help define how quality is measured for AI outputs, ensuring tools are accurate, reliable, and usable in real-world workflows. You will engage directly with engineers across our operating companies to understand workflows, validate solutions, and iterate quickly based on feedback. You may also participate in field-based project hackathons, embedding with teams to identify high-impact opportunities and rapidly prototype solutions that inform platform development. This role requires strong software engineering fundamentals, deep hands-on experience with modern AI tooling, and the ability to operate in a fast-moving environment where both the technology and the product are evolving. You are comfortable with ambiguity, rigorous about output quality, and focused on delivering AI that engineers trust and use. Key Responsibilities AI Application DevelopmentDesign and build AI-powered features using large language models and related toolingDevelop and maintain prompt architectures that drive consistent, high-quality outputsImplement retrieval-augmented generation pipelines using enterprise data sourcesBuild and orchestrate agent-based workflows to automate targeted tasksModel Integration and System BehaviorIntegrate LLM APIs such as Anthropic Claude and OpenAI into production systemsDesign context management strategies to ensure outputs are grounded, relevant, and accurateManage tradeoffs across latency, cost, and performance in AI workflowsContinuously improve system behavior through prompt iteration and architecture refinementPod Collaboration and DeliveryPartner with Software Engineers to integrate AI capabilities into applications, APIs, and user interfacesAlign with the Lead Engineer on technical direction, architecture, and implementation decisionsWork with QA Engineers to define evaluation criteria, testing strategies, and quality thresholds for AI outputsTranslate product requirements into scalable, production-ready AI solutionsEvaluation and Quality OptimizationDefine and implement approaches for evaluating non-deterministic AI outputsBuild test cases, benchmarks, and evaluation pipelines to track output quality over timeIdentify failure modes and iterate on prompts, pipelines, and orchestration logicEnsure consistency and reliability as models, prompts, and data sources evolveContinuous Improvement and InnovationStay current with advancements in LLMs, vector databases, and agent frameworksExperiment with new tools and techniques to improve speed, quality, and capabilityContribute reusable patterns, components, and best practices across pods Skills, Knowledge and Expertise 4+ years of experience in software engineering, applied AI, or machine learning developmentStrong programming skills in Python and/or JavaScriptHands-on experience working with LLM APIs such as Anthropic Claude, OpenAI, or similarExperience designing and implementing prompt architectures and prompt engineering techniquesExperience building retrieval-augmented generation pipelines and working with vector databasesFamiliarity with agent orchestration frameworks and multi-step AI workflowsExperience integrating AI capabilities into applications via APIs and backend systemsStrong understanding of handling structured and unstructured data in AI systemsAbility to evaluate, debug, and improve non-deterministic AI outputsExperience working in a fast-paced, product-oriented development environmentStrong problem-solving skills and ability to operate in ambiguous, evolving contextsAbility to collaborate closely with engineers, product managers, and QA within a pod structureExcellent communication skills and ability to explain technical concepts clearlyCuriosity and willingness to learn domain-specific workflows, particularly within engineering and AEC contexts