Tech Stack

Description

You will design, train, and deploy production-grade machine learning models for document understanding and structured data extraction. Your work powers intelligent document workflows, including layout detection, OCR, retrieval-augmented generation, and AI-first product features for tens of thousands of businesses.

Requirements

• 5+ years of Python experience
• Experience training, fine-tuning, and deploying traditional computer vision models for document intelligence
• Hands-on experience with document understanding frameworks (LayoutLM, Donut, DocFormer) and modern VLMs
• Experience deploying and optimizing models using vLLM, TGI, TensorRT, or ONNX Runtime
• Experience applying LLMs to document intelligence workflows with frontier and open-source models

Responsibilities

• Build and maintain evaluation frameworks for document models, LLMs, OCR, and structured extraction
• Design and curate high-quality datasets for supervised training, fine-tuning, and validation
• Train and fine-tune transformer-based OCR, VLM, layout models, and open-source LLMs for document understanding
• Deploy ML models with modern inference runtimes (vLLM, TGI, TensorRT, ONNX Runtime) and build guardrails
• Develop retrieval and chunking strategies tailored to document structures, optimizing end-to-end RAG pipelines