A leading Industrial Technology company is seeking an AI Engineer for a foundational-stage startup. This platform integrates data from PLM, ERP, and CRM systems to transform historical configurations, invoices, and requirements into a living, validated configuration model for complex industrial machinery.

This role is responsible for building the core intelligence layer.

Technical Scope & Responsibilities

• Intelligence Layer Design: Design and build retrieval and reasoning systems over graph-based product data and cost records.
• AI Feature Delivery: Ship AI features that map incoming requirements to proven configurations, flag risks, and generate reliable engineering outputs.
• Requirements Processing: Extract and normalize requirement signals from CRM and linked systems to guide downstream configuration and validation.
• Model Development: Build AI models and retrieval pipelines that preemptively identify gaps, conflicts, and cost risks, providing actionable recommendations.
• Core AI Services: Implement services utilizing LLMs, embeddings, and graph queries to interpret and act on requirements within a unified product and deal model.
• Orchestration: Architect logic for configuration retrieval, validation runs, and risk escalation. Build pipelines for evaluation, cost estimation, and result publishing.
• RAG/Modeling Advancement: Prototype and harden RAG pipelines using vector databases and knowledge graphs. Focus on prompt/adapter tuning to improve accuracy, latency, and determinism.
• Reliability: Establish evaluation harnesses, golden datasets, and regression tests for models. Instrument latency, cost, and quality budgets with defined SLOs.
• Data Security: Contribute to the architecture for isolation, encryption, access control, and audit, aligning with enterprise requirements.

Required Qualifications

• AI Application Experience: Experience building applications around Large Language Models (LLMs) with a focus on domain-specific knowledge extraction.
• Technical Stack: Strong proficiency in Python and PyTorch.
• Data Expertise: Ability to utilize graph databases or knowledge graphs for contextual data management.
• Infrastructure: Comfort with distributed computing, GPU acceleration, and container orchestration (Docker) for large-scale training or inference.
• Product: Ability to work with ambiguous requirements, iterate quickly, and partner effectively with product managers and systems engineers.

Preferred Qualifications

• Experience deploying AI in enterprise or regulated settings (e.g., manufacturing, aerospace, automotive).
• Knowledge of DevOps practices for reliable model delivery.
• Exposure to HPC or advanced cluster configurations.