Building decision-grade AI systems for complex business problems.

I partner closely with stakeholders to frame ambiguous problems, define decision criteria, and design AI systems that can be trusted in production. My work typically spans problem formulation, modeling and evaluation, and the operational considerations required to deploy and maintain systems over time.

Capabilities

• Decision intelligence systems — framing problems, defining metrics, and building systems that drive action.
• Modeling across modalities — structured data, text, and time-based signals for prediction and forecasting.
• Measurement and causality — experimentation, MMM, and causal approaches to quantify impact.
• Production-minded evaluation — calibration, monitoring, and failure-mode analysis for reliable performance.

Credibility

I build and evaluate AI systems deployed in real operating environments, where decisions carry financial, operational, and regulatory consequences.

My experience spans structured and unstructured data at scale, combining classical modeling approaches with modern GenAI / LLM-based systems, including retrieval-augmented pipelines and decision support applications.

A core focus of my work is rigorous evaluation—calibration, robustness testing, failure-mode analysis, and measurement, ensuring systems remain reliable, interpretable, and trustworthy in production, particularly in regulated environments such as healthcare.

Selected public work and background:

• GitHub
• LinkedIn

Contact

If you’d like to discuss a role, consulting engagement, or collaboration, email me.

Email me (I typically respond within 1–2 business days.)