Work Experience
The full timeline.
Software Engineer II
I lead disaster recovery for Fabric Maps. That means analyzing how a failover would ripple across every workstream, building mitigations that match the wider platform's approach, validating resilience in pre-production rings, and expanding the metrics and telemetry we track once something ships to production. I also led OAP, a cross-team security initiative that started with a customer's data-boundary requirements and turned them into concrete technical scope, then drove the design and build together with the platform team and every other dependent workload involved. On top of that, I've become the team's go-to person for AI: I built a Fabric Maps MCP server, an Azure Maps assistant that turns natural-language prompts into map visualizations, tools for agentic repository onboarding, and a set of reusable plugins and skills, and I run sessions to share what I'm learning with the rest of the team.
Software Engineer
I owned the Tileset Job API from the ground up, the system Fabric Maps relies on for large-scale geospatial ingestion and map-tile generation. That meant designing the architecture, building the .NET backend and its frontend integration, and handling validation, testing, and telemetry myself, then iterating as real feedback came in from stakeholders. I also picked up a good share of the team's security and reliability work during this period, including migrating our Dev-Ops pipelines to YAML and building a PowerBI-driven load-testing system.
Computer Vision Researcher
Built a TensorFlow-based license-plate recognition model for a campus security system, hand-annotating 1,000 training images to reach 65% accuracy. Also fine-tuned a BERT model to process and query legal documents.