We're building production AI systems that automate document-heavy and knowledge-heavy workflows across the business. As an AI Engineer, you'll own intelligent document processing pipelines (OCR + LLM), build copilot-style assistants on top of enterprise data, and ship everything end-to-end on Databricks and Azure.

• Intelligent Document Processing. Design and ship OCR + LLM pipelines that extract, classify, and validate information from scanned documents and PDFs. You'll own the full loop — from layout-aware parsing to entity extraction, confidence scoring, and feedback-driven improvement.
• Conversational AI on enterprise data. Build copilot-style assistants and query interfaces over our Databricks data products, with proper authentication, role-based access, and grounded responses.
• End-to-end delivery on Databricks and Azure. PySpark pipelines, Delta Lake, MLflow for experiment tracking and model registry, batch and real-time inference. Deploy on Azure with Azure OpenAI, Azure AI services, ADLS, Key Vault, and proper monitoring. Set up CI/CD and MLOps practices that make releases boring.
• Reusable foundations. Prompt patterns, evaluation harnesses, pipeline components, model templates — the assets that let the team move faster on the next use case.

Must have:

• Strong Python and PySpark skills, with real experience shipping ML or data products to production
• Solid foundation in classical ML (classification, clustering, similarity, anomaly detection) — you know when not to reach for an LLM
• Hands-on experience with LLMs in production: RAG, extraction, evaluation, prompt design
• Databricks delivery experience (Delta Lake, MLflow, performance tuning)
• Comfortable on Azure: identity, secrets management, networking, and the AI/data services
• Engineering discipline: Git, code review, testing, clear documentation
• Fluent with AI coding assistants (Cursor, Claude Code, Copilot, or similar) as part of your daily workflow
• Able to work directly with business stakeholders, debug across the stack, and explain trade-offs clearly.

Nice to have:

• Experience with OCR pipelines and document AI in production
• Exposure to LLM evaluation frameworks and automated testing for AI systems

• Experience with Responsible AI, security controls, and enterprise compliance

  controls, and enterprise compliance