Skip to main content

HI, MY NAME IS

James Stopford

A developer who does data.

I'm a Senior Data Platform Engineer & Tech Lead at Amsted Digital Solutions, where I turn high-volume event streams into real-time intelligence for the rail industry.
01.

About

I'm a developer who does data.

I build the systems that turn high-volume event streams into something people can actually use β€” the backends that ingest millions of events a day and keep them consistent, and the data platforms that make all of it queryable and trustworthy. I've stood up a production data lake from scratch, solo, and own it end to end.

Based outside Philadelphia, PA πŸ‡ΊπŸ‡Έ

Outside work you'll usually find me gaming, powerlifting, or reading β€” mostly sci-fi and non-fiction.

WORKS WITH

  • Azure
  • Iceberg / Trino
  • Databricks
  • Cosmos DB
  • Terraform
  • Docker & Kubernetes
  • C# / .NET
  • Python
02.

Experience

Senior Data Platform Engineer / Tech Lead Β· Amsted Digital Solutions

Oct 2025 β€” Present

I lead the build-out of Amsted Digital Solutions' data platform: a production Azure data lakehouse on ADLS Gen2, fully managed with Terraform, powering self-service analytics for rail fleet intelligence.

  • β–Ή Designed and shipped Bronze/Silver/Gold pipelines on ADLS Gen2 + Iceberg for high-volume device telemetry and operational data, with time-partitioned Parquet landing and private endpoint security.
  • β–Ή Stood up Apache Polaris (Iceberg REST catalog) and Trino on AKS, and deployed Apache Superset as the primary self-service BI and SQL layer.
  • β–Ή Established the platform's foundations: schema registry, contract-first modeling, Architecture Decision Records, security posture, and an RBAC/governance framework.
Azure (ADLS, AKS) Terraform Iceberg / Trino / Polaris Databricks / Spark Kubernetes Python

Tech Lead Β· Amsted Digital Solutions

May 2020 β€” Oct 2025

I led the core rail telematics and asset-management platforms β€” event-driven systems that turn a firehose of railcar signals into real-time fleet visibility across thousands of assets.

  • β–Ή The heart of it is an event-sourced asset-management platform with real-time profiling and predictive sequencing, fed by pipelines that process 4M+ events a day with exactly-once guarantees.
  • β–Ή Just as important was the work outside the code: partnering with hardware engineering and rail ops to nail down requirements and prove things out in production.
C# / .NET Azure Cosmos DB SQL Event-Driven Architecture MQTT REST / Graph APIs

Lead Software Developer Β· Atos

Aug 2013 β€” May 2020

Led a team of five to build the vulnerability and threat management platform that became the foundation of an enterprise cyber-security offering β€” including a large-scale threat intelligence system that pulled, analyzed, and securely delivered content from across the public internet and dark web.

C# / .NET SQL Big Data Security Platforms Python Team Leadership

Software Developer Β· Siemens IT Solutions and Services

Oct 2008 β€” Aug 2013

The sole developer embedded in an operational account management team, where I hunted for high-impact automation and integration opportunities and built the production tooling to deliver on them β€” custom integrations and automation that measurably improved efficiency for large enterprise customers.

Process Automation Systems Integration C# / .NET SQL
03.

Projects

AZURE LAKEHOUSE

Self-Service Analytics Platform

A production Azure lakehouse built and owned solo, end to end. Bronze/Silver/Gold pipelines on ADLS Gen2 + Iceberg ingesting high-volume device telemetry, with an Iceberg REST catalog (Polaris) and Trino query engine running on AKS, all managed through Terraform. Superset on top as the self-service BI and SQL layer. Includes schema-registry-driven contract validation, MERGE-based idempotency, and a compaction/maintenance job for table health.

Icebergβ€’Trinoβ€’Polarisβ€’Databricks/Sparkβ€’Terraformβ€’Supersetβ€’Azure
Superset Plugin

deck.gl Trip-Path Visualization

A custom Superset chart plugin for rendering GPS trip routes from GeoJSON-native data β€” a PathLayer for the route with IconLayer directional arrows showing travel direction. Built because off-the-shelf options couldn't render route geometry without coordinate explosion. Packaged as a standard Superset plugin (plugin-chart-trip-paths) against the 6.0 plugin API.

Reactβ€’deck.glβ€’Supersetβ€’TypeScriptβ€’GeoJSON
DATA GOVERNANCE

A Lightweight Data Product Request Process

Designed and ran a formal intake process for a self-service data platform: a request template, a Kanban workflow, and clear handoffs from upstream governance to delivery. Turned ad-hoc "can you build me a table" asks into versioned, traceable data products with defined ownership β€” the connective tissue that keeps a one-person platform from becoming a bottleneck.

Data Governanceβ€’Process Designβ€’Stakeholder Management
THIS WEBSITE

jamesstopford.com

The site you're looking at. Clean, accessible, and designed for rapid iteration. Built as a simple static site with vanilla HTML, CSS and JS.

HTMLβ€’Tailwind (CDN)β€’Vanilla JSβ€’Static
04.

What's Next?

I'm currently heads-down on rail telematics and industrial data systems, but I'm always up for a good conversation about data platforms, distributed systems, or making high-volume data behave at scale.

Say Hello