Leading the design and development of Oasis, Shepherd’s next-generation pricing engine — a Python-based service that replaces legacy TypeScript infrastructure with a modern compute layer built on FastAPI, Pydantic, and Polars.

Re-architecting the pricing system around a semantic, domain-driven model of pricing operations, transforming a legacy, code-centric implementation into a composable, auditable execution framework aligned with underwriting and actuarial concepts.

Pioneering a spec-driven, AI-augmented development methodology in which structured, machine-readable design artifacts (architecture, API contracts, pricing workflows, and test strategies) serve as the authoritative system definition. These specifications enable AI-assisted design exploration, implementation, and validation, effectively shifting development from code-first to design-as-execution, accelerating delivery while maintaining strong engineering rigor and correctness.

Introduced a new PDF export pipeline into the product, using headless Chrome, that delivered critical compliance related functionality for a large customer on a tight timeline.

Developed an automated solution to continuously validate completeness of the product's existing audit logs against database transactions via a Kafka-based Change Data Capture (CDC) pipeline.

• Achieved zero reported defects with respect to the commanding and controlling of our first commercial satellite launch and orbit raise (the most critical phase of the satellite's life). I had initiated the development of the mission critical command and control web application development that manages sending commands to spacecraft. This included selecting the tech stack, bootstrapping the development, mentoring full-stack engineers, significant development contributions, and being responsible for the overall quality.

• Ensured satellite anomaly visibility across a large number of satellites by rearchitecting a violations monitoring subsystem for near real time visibility of anomalies across an entire fleet. This included architecture to achieve horizontal scalability, all backend development and significant guidance to the UI design.

• Evaluated, benchmarked and integrated the TimescaleDB timeseries relational database for high frequency telemetry ingestion from satellites. This included developing tooling for automated schema definitions.

Led a small, empowered, high-performing team of engineers to build out an in-house Laboratory Information Management System (LIMS) supporting the unique data tracking requirements of Aether's high-throughput enzyme engineering factory.

Defined architecture, prototyped technologies and developed scalable solutions for a new Traffic Performance Analytics platform supporting key new multi-tenancy Iteris products such as SPM (Signal Performance Measurement) and re-engineered ones such as ClearGuide (Traffic Analytics). Defined a micro-service strategy, built a central authentication solution, built out a large scale analytics engine using "serverless" technologies, big data aggregation.

Also led the development team in adopting best practices and driving change in the areas of automated testing, code reviews, continuous integration and delivery, and release planning.