GenAI Engineer, Document Intelligence (Multimodal)

MaintainX; United States; On-site; USD 140,000 - 200,000; Full time

Job summary

MaintainX seeks a skilled GenAI specialist to lead the generation pipeline within Document Intelligence. You will design prompts, establish per-entity schemas, and build eval datasets to ensure high-quality, trustworthy outputs from multimodal inputs such as video, PDFs, and OCR.

You will collaborate with the Engineering Lead and the Processing side of Document Intelligence, shaping model selection, budgets, and deployment strategies to balance quality, speed, and cost.

Qualifications

• Strong GenAI craft: prompt engineering, structured output, tool-use.
• Real eval discipline: built datasets and rubrics, measured quality.
• Shipped LLM features into production services.
• Comfort working with multimodal inputs - video, PDF, audio, image to text/structured output.

Responsibilities

• Design and iterate prompt recipes (system, few-shot, context) for generation.
• Define per-entity target schemas and validators for generated outputs.
• Build generation-quality eval datasets and rubrics, offline and online.
• Assemble multimodal context windows from keyframes, transcripts, and OCR.
• Choose model and token budget per recipe based on quality, cost and latency.

Skills

GenAI craft

Eval discipline

Production ML features

Multimodal inputs

Job description

MaintainX seeks a skilled GenAI specialist to lead the generation pipeline within Document Intelligence. You will design prompts, establish per-entity schemas, and build eval datasets to ensure high-quality, trustworthy outputs from multimodal inputs such as video, PDFs, and OCR.

You will collaborate with the Engineering Lead and the Processing side of Document Intelligence, shaping model selection, budgets, and deployment strategies to balance quality, speed, and cost.