Description

You will design and build end-to-end text extraction pipelines for policy, regulatory, fintech, and healthcare documents.

Responsibilities

• Design and build end-to-end text extraction pipelines for complex document types.
• Extract key entities and structure policy clauses and obligations.
• Fine-tune BERT and RoBERTa for NER, text classification, and relation extraction tasks.
• Leverage LLM APIs using prompt engineering, tool/function calling, and structured output extraction.
• Build scalable Python pipelines for high-volume processing of PDF, DOCX, and HTML files.
• Define and enforce JSON schemas to ensure outputs are compatible with knowledge graph ingestion.
• Evaluate model performance and implement feedback loops to improve extraction quality.

Required Skills

• 5+ years of hands-on NLP engineering in production pipelines.
• Python.
• NLP (spaCy, NLTK).
• HuggingFace Transformers.
• Deep Learning.
• LLM API Integration.
• Data Pipeline development.
• JSON Schema.
• Pydantic.
• Any Graduate degree.

Preferred Skills

• Experience with legal, regulatory, or policy documents.
• Familiarity with knowledge graphs or graph databases like Neo4j or RDF.
• Document parsing tools including pdfplumber, Docling, or Apache Tika.
• Domain knowledge in fintech or healthcare NLP.
• Exposure to information extraction benchmarks such as CoNLL, DocRED, or SciERC.

Key Skills

Education

Any Graduate