Graeme Watson

Lead data scientist and applied AI technical lead delivering GenAI, analytics, and decision-support systems across Databricks and Azure. Own the path from ambiguous business problems through discovery, scoping, solution design, implementation, evaluation, and adoption.

• Built and deployed an internal Databricks RAG chatbot using LangChain, React, Databricks Apps, retrieval evaluation, guardrails, and feedback loops; established a reusable delivery pattern for future GenAI applications.
• Lead cross-functional AI and analytics initiatives from ambiguous problem definition through technical design, implementation, deployment, and adoption.
• Led AMI operational monitoring delivery in partnership with Gas Measurement Integrity, including ingestion design and endpoint monitoring dashboards used by key groups across the company.
• Improved production anomaly detection by replacing weaker ML-style approaches with tested SQL-based rules in Databricks, increasing verified detections and operational reliability.
• Led root-cause analysis, stakeholder coordination, vendor management, and rollout planning for a production workflow remediation, reducing duplicate task creation.
• Evaluated AI coding tools and trained the back-end team on AI-assisted development workflows, helping shape internal guidance and adoption.

Built analytical tools and decision-support models for operational planning, resource allocation, and business improvement, combining technical delivery with business-facing problem solving.

• Built and deployed a Python-based linear optimization tool for contractor capacity planning and budget allocation across an approximately $2M portfolio, reducing budget variance by about $200K and improving the monthly planning cycle.
• Developed and deployed forecasting models for field staffing and resource planning, improving allocation accuracy by approximately 15% and supporting better operational decision-making.
• Translated operational planning problems into analytical tools, recommendations, and decision-support workflows.
• Partnered with leadership on operational and analytical initiatives that connected business priorities with practical technical execution.

• Built executive-facing Power BI dashboards translating operational data into decision-ready KPIs and performance insights.
• Supported management reporting and operational visibility through SQL-, Python-, and BI-based analysis.

• Supported customer discovery, market research, MVP definition, and business planning for a two-sided marketplace.