Principal AI Application Architect

Job Description — AI Solutions Architect / Senior AI Engineer

Long Term Contract

Vacaville, CA

Overview

We are seeking an experienced AI Solutions Architect / Senior AI Engineer to lead the design and development of an end-to-end AI application on our on-premises GPU infrastructure (4 × NVIDIA L4 GPUs). This role will collaborate closely with the Process Engineering and Automation teams to modernize and automate document-processing workflows, demonstrate the power of AI, and deliver high-impact business transformation.

Key Responsibilities

1. AI System Architecture & Development

• Architect and build a local, GPU-optimized AI platform for document intelligence, workflow automation, and RAG/agentic systems.
• Develop scalable pipelines for:
• PDF → Structured Data (Excel/CSV)
• Extraction of tables, diagrams, flows, and structured text from vendor documents.
• OCR, vision-based parsing, and layout-aware models.
• Implement RAG pipelines, vector search, and agent-based AI workflows.

1. Document Automation & Workflow Integration

• Automate conversion of vendor PDFs into internal formats:
  SFA → PDTT → SDTT → downstream systems.
• Build transformation logic to populate new columns, normalize schemas, and validate data.
• Integrate AI-generated outputs into business workflows with high accuracy and auditability.

1. Application Development

• Build a user-facing AI application (local/on-prem) to:
• Upload vendor documents.
• View extracted and converted data.
• Approve/modify outputs.
• Trigger automated workflows to other teams (Process, Automation, QA, etc.).

1. Performance Optimization

• Optimize model inference, batching, and pre/post-processing for the onsite GPU cluster (L4 GPUs).
• Improve transfer efficiency, memory use, I/O, and system throughput.

1. Collaboration & Cross-Functional Work

• Work closely with Process Engineering, Automation, and Business Enablement teams to understand workflows and constraints.
• Translate business requirements into technical specifications and deliver a measurable MVP.

1. Advanced AI Capabilities

• Build and maintain:
• Vision AI models (diagrams, engineering flows, layout extraction).
• LLM agent workflows (multi-step reasoning, tool use).
• RAG systems with vector DBs (FAISS/MilvChroma).
• Text-to-SQL pipelines for automatic query generation.
• Ensure outputs are explainable, verifiable, and production-grade.

Qualifications

Must Have

• 5+ years in AI/ML, NLP, or Data Engineering (or equivalent senior experience).
• Strong experience with:
• LLMs (open-source, fine-tuning, quantization)
• RAG systems
• Agent frameworks (LangChain, LlamaIndex, etc.)
• OCR, document extraction, or layout models
• Python, PyTorch/TensorFlow
• Hands-on experience deploying AI on on-prem GPU infrastructure.
• Experience building end-to-end pipelines (ingest → extract → transform → deliver).

Nice to Have

• Experience with PDF parsing libraries (PDFPlumber, Camelot, Tabula).
• Experience with workflow orchestration tools (Airflow, Prefect, Temporal).
• Experience with manufacturing or engineering documentation.
• Understanding of SFA/PDTT/SDTT or similar business process flows.
• Experience converting engineering diagrams or flowcharts using Vision AI.

What You Will Deliver (MVP)

• A working application that:
• Ingests vendor PDFs.
• Converts them into structured Excel/SFA-ready data.
• Automates transformations and downstream workflows.
• Showcases “the power of AI” through RAG, agents, and intelligent document processing.