AI Integration Engineer

Loenbro•Gilbert, AZ

6h•Hybrid

About The Position

Build and deploy practical, secure AI solutions that accelerate construction operations. This role blends prompt engineering, MLOps, API integration, and user-facing tooling to deliver lightweight, production AI agents and automations that reduce manual work and improve field productivity.

Requirements

Bachelors in Computer Science, Data Science, Engineering, or equivalent.
5+ years in data engineering, software engineering, or related technical roles.
3+ years hands-on AI/ML development and deployment.
1+ year deploying agentic workflows or AI agents in production.
Proven experience integrating AI into enterprise systems via APIs/microservices.
Strong Python skills; experience with LLM APIs, vector DBs, containerization (Docker/K8s), and cloud ML platforms.
Demonstrated experience building and deploying agentic workflows or AI agents in a production environment or a significant internal project, or hands on portfolio work.
Knowledge of data security, RBAC, and compliance in enterprise environments.

Nice To Haves

Experience in construction or heavy industry workflows and systems (ERP, Procore, Viewpoint).
Familiarity with MLOps tools (MLflow, TFX, Kubeflow) and observability platforms.
Hands-on experience working within one or more major LLM platforms — OpenAI, Azure OpenAI, or Anthropic.
Certifications in cloud ML (AWS/Azure/Google) or security.

Responsibilities

Design, build, deploy: deliver lightweight AI tools (chatbots, assistants, agentic workflows) that solve high value operational problems.
Partner with operations: identify repetitive, high friction processes with field teams and convert them into AI workflows.
Prompt engineering & validation: develop prompt strategies, test for accuracy, and implement guardrails to reduce hallucinations.
Prototype & deploy: move prototypes from LLM APIs and automation platforms (Zapier, Power Automate, custom scripts) into secure, monitored production services.
API integrations: integrate models with internal systems and third-party services via microservices and secure APIs.
MLOps & monitoring: implement CI/CD for models, observability for model drift, and performance dashboards for adoption metrics. Implement CI/CD for AI workflows, monitor response quality and prompt performance over time, and build dashboards that track adoption, usage patterns, and output accuracy.
User interfaces: build simple UIs or dashboards for non technical users to interact with AI tools.
Governance & security: enforce RBAC, data handling policies, and compliance controls for sensitive data.
Development Lifecycle: use deployment best practices, including application lifecycle management and change control.
Documentation & training: document workflows and facilitate training of end users; hand off to IT for long term support.
Roadmap contribution: help evolve standalone tools into integrated AI architecture aligned with enterprise strategy.