Overview

As a Senior AI Scientist on BI, you will own AI/ML solutions end to end, from data curation through evaluation to production, applied directly to BI's reporting and financial-analysis surfaces.

Responsibilities

• Own delivery of end-to-end AI/ML solutions for BI surfaces, spanning modeling, evaluation, and production integration.
• Select appropriate algorithms and approaches for financial and reporting problems, explain trade-offs, and identify common failure modes such as hallucination, bias, and drift.
• Curate task-specific datasets from Intuit's proprietary financial data, with data governance and privacy built in, and build evaluation benchmarks specific to BI use cases, such as the accuracy of generated financial commentary or KPI explanations.
• Design AI-native features that balance latency, cost, and quality, including features embedded in agentic pipelines such as Omni conversational BI, and design evaluation approaches for non-deterministic agentic flows.
• Track emerging research and apply novel ML methods, including causal inference, time-series forecasting, and NLP/LLMs, to BI problems such as anomaly detection in financial statements and multi-entity consolidation insights.
• Use AI agents as part of your own science workflow, including experiment orchestration, hyperparameter search, and simulation-based validation, and help set standards for which BI science tasks should be agent-driven.
• Partner with BI product, design, and engineering to shape what's worth building, grounded in usage and feedback signals from reporting surfaces.
• Apply responsible AI practices given that BI surfaces present financial figures directly to accountants and business owners, and mentor others on the same.

Qualifications

• MS or PhD in Computer Science, Statistics, Applied Mathematics, Operations Research, or a related field, or equivalent practical experience.
• 4+ years of applied ML or data science experience, ideally including work with financial, accounting, or structured tabular and time-series data.
• Hands-on experience with LLMs and agentic AI systems, including prompting, fine-tuning, and evaluation of agent pipelines, alongside classical ML methods such as supervised and unsupervised learning, time-series forecasting, and anomaly detection.
• Proficient in Python and SQL, comfortable in a Linux environment, and experienced with large-scale distributed data systems such as Spark or Hive; familiarity with OLAP engines such as StarRocks is a plus.
• Experience building end-to-end, reusable ML pipelines, from data acquisition and ETL through model deployment and monitoring.
• Strong written and verbal communication skills, with the ability to explain technical and financial concepts to both technical and non-technical audiences.