AI Lead

Job Description

Must have skillsets and experience 1. Retrieval Augmented Generation & IDP Building hybrid search (BM25 + embeddings) and re ranking pipelines; tuning recall/precision for unstructured card statement text. Designing and orchestrating Expert Agents (Extraction, Pricing, Validation) with LangChain/Snowpark task graphs. 2. Applied MLOps inside a regulated enclave GitHub Actions / Terraform CI CD, model versioning in MLflow or Snowflake, drift & cost dashboards in Prometheus / Tableau. Writing unit + synthetic statement tests that gate promotions, and rolling models forward under strict rollback guarantees. 3. Payment domain compliance & performance PCI DSS data handling patterns, PII masking, SOX grade lineage that captures embeddings, prompts and responses for 18 months. Meeting hard SLAs ( 5 min end to end ingestion; 90 s quote recompute) and sizing infra to keep GPU/credit burn visible to FinOps. 4. Full stack data & model engineering Deep Python/SQL plus either Snowflake VECTOR_SEARCH or Databricks Lakehouse vector indexes, and comfort switching between them. Experience wiring OCR at scale (Tesseract/Textract on UDFs or Spark) and streaming outputs into secure warehouses. 5. Collaboration in a quad team RACI model Willingness to act as the Responsible engineer while an AI Architect is Accountable, partnering tightly with the Business SME (fee taxonomy) and MLQA Engineer (accuracy evidence). Ability to translate compliance or finance feedback directly into backlog tasks without waiting for long managerial chains. 6. Continuous learning mindset Designing active learning loops that detect uncertainty, call for human labels and retrain nightly, keeping extraction and pricing accuracy high as statement formats evolve