
James Y
Bulk AI Data and Compute Pipeline Specialist
Skills

See my services

Portfolio
Work experience
Founder & Lead Systems Engineer
Google Developers • Self-employed
Apr 2025 - Present • 1 yr 3 mos
Designed and deployed automated multimodal AI data extraction pipelines that convert unstructured PDFs, invoices, and bank statements into structured datasets with high schema accuracy. Built serverless batch inference workflows on cloud infrastructure (Azure / Lambda GPUs), reducing processing turnaround times by over 90% compared to manual data entry. Integrated automated Pydantic schema validation and multi-format export engines delivering data in Excel (.xlsx), CSV, JSON, and SQL formats.
Principal Infrastructure Developer
NVIDIA • Self-employed
Aug 2023 - Sep 2024 • 1 yr 1 mo
Built an open-source AI orchestration framework featuring intelligent multi-model task decomposition and automated document parsing. Implemented structured JSON output validation and data cleaning pipelines to handle high-volume text and visual inputs reliably. Optimized API gateway throughput and asynchronous execution protocols to handle multi-file batch jobs efficiently.
Database Optimization Consultant
Vacasa • Freelance
Jan 2021 - Dec 2021 • 11 mos
Identified performance bottlenecks in large-dataset queries and proposed more efficient retrieval strategies to improve system scalability. Focused on closing the gap between functional SQL and high-performance SQL for large-scale production environments.