Lead Architectural AI Engineer

Our client is a custom software development company out of Grand Rapids, Michigan, that exists to make software that humans actually like to use.

Type: P/T Contract 20‑40 hours/week to start
Duration: 2 months to start

Location:

West MI (Prefer hybrid, will look at remote)

The Technical AI Enablement Engineer leads the architectural implementation of AI across the enterprise. Unlike a traditional software engineer, your focus is specifically on the orchestration layer building RAG pipelines, managing API integrations, and developing middleware that enables business units to leverage LLMs safely, efficiently, and at scale.

• Design and deploy AI agents using frameworks such as Lang Chain, Llama Index, or AutoGPT to automate complex, multi‑step business logic.
• Build and maintain Retrieval‑Augmented Generation pipelines, including managing vector databases (e.g., Pinecone, Weaviate, or Milvus) and document ingestion workflows.
• Develop robust Python or Node.js middleware to connect frontier models (OpenAI, Anthropic, Gemini) with internal legacy databases and CRM systems.
• Implement "LLM‑as‑a‑judge" frameworks and automated testing suites to measure model accuracy, latency, and hallucination rates in production.
• Establish technical guardrails for PII masking, prompt injection mitigation, and token cost optimization across all internal applications.
• Guide the selection of hosting environments (e.g., AWS Bedrock, Azure AI Studio) and manage model versioning and deployment cycles.

• Professional proficiency in Python (specifically for data processing and AI backends) and Type Script/JavaScript.
• Deep experience with Lang Graph or Haystack for building stateful, multi‑agent workflows.
• Strong SQL skills and experience working with vector databases and unstructured data ETL processes.
• Working knowledge of RESTful API design, Webhooks, and authentication protocols (OAuth, API Keys).
• Experience with Git/Git Hub, containerization (Docker), and CI/CD pipelines for AI‑powered applications.
• Prompt Engineering (Technical):
  Mastery of advanced techniques including ReAct prompting, chain‑of‑thought, and algorithmic prompt optimization.

• 4+ years in Software Engineering, Data Engineering, or Solutions Architecture.
• 2+ years of hands‑on experience building and deploying LLM‑based applications in a production environment.
• Education:
  
  BS/MS in Computer Science, Data Science, or a related technical field.
• Portfolio:
  Ability to demonstrate a Git Hub repository or technical project involving autonomous agents or a complex RAG system.