Senior AI Engineer – Agentic OCR & Document Intelligence

We are hiring a Senior AI Engineer to design and build a next-generation agentic OCR and document processing system. You will own the end-to-end development of a multi-agent pipeline that extracts, validates, and delivers structured data from business documents using AI.

This role goes beyond traditional OCR — you will architect intelligent, self-healing systems that improve accuracy and reduce manual review through targeted AI workflows.

Key Responsibilities

• Design and build a multi-layer AI agent pipeline for document processing
• Develop OCR extraction, classification, and schema mapping systems
• Implement confidence scoring and automated validation logic
• Build self-healing workflows for low-confidence data extraction
• Orchestrate workflows using LangGraph and AWS Step Functions
• Develop and integrate MCP-based AI tools and APIs
• Optimize OCR engines for accuracy, cost, and performance
• Build scalable infrastructure using AWS, Docker, and Terraform
• Ensure system reliability with monitoring, testing, and observabilityRequirements
• 5+ years of Python development (production-level)

Good to Have

• Computer vision / OpenCV experience
• Document classification & schema mapping
• Human-in-the-loop (HITL) systems
• LLM observability tools (LangSmith, Arize, etc.)
• Experience with logistics, manufacturing, or document-heavy industries
• 2+ years working with LLMs and AI systems
• Experience with LangChain / LangGraph or similar frameworks
• Strong hands-on experience with AWS (Bedrock, Lambda, S3, ECS, Step Functions)
• Solid understanding of OCR and document processing systems
• Experience building APIs (FastAPI) and working with PostgreSQL / OpenSearch
• Familiarity with Docker, Terraform, and cloud infrastructure
• Strong English communication skills (B2+)

Application Question(s):

• What is your current and expected salary?
• Can you join on an immediate basis?

Work Location: In person