On Prem - AI Engineer
Core
Design, secure, and expand an on-premises LLM environment integrated with a manufacturing ERP to automate reporting, catalog management, and operational queries.
Role type
Senior AI Engineer (On-Premises LLM & ERP Integration)
Builds
Closed-LAN LLM infrastructure, SQL-based RAG pipelines, automated report libraries, and catalog/OEM data connectors.
Domain
Manufacturing (Automotive Aftermarket/OEM) + Enterprise AI
Required skills
Open-weight LLM deployment, NVIDIA hardware tuning, SQL (mature ERP schemas), Python, RAG pipeline construction, LAN security architecture, ERP integration (Macola)
Preferred skills
Manufacturing domain knowledge, legacy report modernization, multi-GPU workstation builds, high-VRAM inference optimization
Technologies
NVIDIA Spark, vLLM, TensorRT-LLM, Ollama, Macola, Open WebUI, Python, SQL, Excel
Responsibilities
Own and secure the closed, on-premises LLM environment on NVIDIA hardware; Connect AI capabilities to the Macola ERP using SQL and RAG; Build a reliable report library to replace legacy reports; Develop Excel-to-Macola loaders and catalog connectors; Deploy the system on a LAN-only basis; Present weekly demonstrations to leadership; Document the system and plan the hardware upgrade.
Seniority
Senior, hands-on IC
