remote
AI/ML Engineer - SecHorizon Technologies Pvt. Ltd
ML Engineer
AI/ML Engineer focused on large‑scale document intelligence, mastering Azure Document Intelligence, table extraction with Camelot/Tabula, and custom transformer models like Donut and LayoutLM, with end‑to‑end production deployment and rigorous evaluation.
About the role
Key Responsibilities
- Design, develop, and deploy scalable document extraction pipelines using Azure Document Intelligence at production scale.
- Implement advanced table extraction for complex layouts (merged cells, borderless tables, multi‑column) with Camelot, Tabula, or equivalent tools.
- Build and fine‑tune custom ML models for layout classification, field extraction, and document understanding (Donut, LayoutLM).
- Establish robust evaluation frameworks: create ground truth datasets, compute per‑field precision/recall, and drive continuous quality improvement.
- Collaborate with data scientists and product teams to integrate extraction outputs into downstream applications.
Requirements
- Hands‑on production experience with Azure Document Intelligence and deep understanding of its capabilities and limits.
- Proficiency in table extraction libraries (Camelot, Tabula) and experience handling complex document structures.
- Strong background in training and deploying transformer‑based document models (Donut, LayoutLM).
- Solid knowledge of evaluation metrics, dataset creation, and systematic quality measurement.
- Excellent problem‑solving skills and ability to work cross‑functionally in a fast‑paced environment.