Build production AI systems for finance teams, including customer-specific anomaly-detection models, grounded natural-language generation, agentic workflows, and managed LLM integrations. Develop Python services with warehouse access and customer-hosted deployments. Implement verification, human-confirmation checkpoints, anti-leakage controls, and CI evaluation for grounding and hallucination. Use customer feedback to improve models and systems while collaborating with AI and evaluation engineers.

Design and build deterministic, auditable AI engines and conversational agents for finance platforms. Lead architecture across RAG, agent orchestration, structured outputs, grounding, evaluation, verification, and human-in-the-loop workflows. Develop production Python services with SQL and vector retrieval, support multi-cloud and on-premises deployments, mentor engineers, conduct design reviews, and communicate architecture decisions to technical and business stakeholders.

Build and own production-grade data pipelines, source connectors, warehouse transformations, curated data models, reconciliation, and lineage for an AI-native finance platform. Develop dbt projects, orchestrate reliable and rerunnable workflows with Airflow or Dagster, tune warehouse performance, integrate messy enterprise systems, and ensure data correctness, auditability, and tenant isolation. Collaborate with backend and AI teams, establish data contracts, review work, and mentor engineers.