AI Outlook Sales Agent
An AI agent living in Outlook that reads incoming sales orders, checks status, and answers product questions against live ERP data.
- Model Context Protocol
- Microsoft Dynamics 365
- Azure
- AI agents
Read the write-up
Projects
Each write-up includes the parts I found hard, which are usually the parts I learned the most from.
An AI agent living in Outlook that reads incoming sales orders, checks status, and answers product questions against live ERP data.
Read the write-up
Model Context Protocol servers exposing live Dynamics 365 Finance & Operations data (customers, products, orders, inventory) to AI agents.
Read the write-up
An end-to-end Sales Orders and Sales Returns workflow built on Celigo, connecting a Shopify storefront to Microsoft Dynamics 365 Business Central.
Read the write-up
A research project with AstraZeneca turning unstructured manufacturing equipment data into a queryable model the business could act on.
Read the write-up
Architecture and provisioning for AI user interfaces backed by agentic logic, built so one tenant's data and prompts can't reach another's.
Read the write-up
A sandboxed AI chatbot provisioned on demand into an e-commerce site to answer product questions, with infrastructure that only exists while it's in use.
Read the write-up
Boomi integration development supporting a global telecom supply chain organization's ERP implementation.
Read the write-up
Guidepath's B2B storefront for its own ERP and WMS modules, built to distribute software and capture business leads on the same path.
Read the write-up
Power Automate flows that sync Azure DevOps work items into Microsoft Planner, turning a draft workflow into something the team actually runs on.
Read the write-up
A research project with AstraZeneca replacing paper-based QC logbooks with a digital system, reducing manual errors and accelerating audits.
Read the write-up
Cleaning a Center for Civic Education order history, then building a national heatmap plus bar and line charts to show who their educational materials were reaching.
Read the write-up
Python tooling written with the USI IT department to help diagnose and patch network vulnerabilities across university systems.
Read the write-up