Exavalu is looking to solve the problem of document ingestion for P&C Specialty Lines using AI and machine learning
Requirements
- AWS Bedrock
- AWS Lambda
- S3
- Step functions
- Anthropic Claude
- Python
- Pinecone
- FAISS
- OpenSearch
- Chroma
Responsibilities
- Build PDF document ingestion pipelines for P&C Specialty Lines and identity documents (using OCR/NLP)
- Hands-on experience with AWS Bedrock, AWS Lambda, S3, Step functions and Anthropic Claude for LLM-driven document understanding
- Implement embeddings pipelines and store/search in vector databases (Pinecone, FAISS, OpenSearch, Chroma etc.)
- Deep understanding of LLMs, embeddings, prompt engineering, and model lifecycle management
- Knowledge of document parsing frameworks (Amazon Textract, PyPDF2, PDFPlumber, Tesseract)
- Build PDF document ingestion pipelines for P&C Specialty Lines and identity documents (using OCR/NLP)
- Expertise in Python for AI/ML and serverless development
Other
- Background in P&C insurance technology or financial services
- Diversity and inclusion
- Part-time work, working outside normal 9-5 business hours or working remotely
- Welcome back program to help people get back to mainstream after a long break due to health or family reasons
- Commitment to building a diverse and inclusive workforce
