Joined team building back office finance systems in Java and Python (Airflow)

Led on strategy for sharing finance data for analytics purposes

Championed unifying team’s approach to Terraform and infrastructure

Hired initially to lead Backend Chapter and drive through a “Tech Transformation” strategy to break a Ruby monolith into Python FastAPI services for a new Svelte app.

Worked with management to prove improvements in velocity and continuous delivery due to transformed architectures.

Tech: Python (FastAPI), TypeScript (Svelte), PostgreSQL, DocumentDB, data pipelines, Ruby (Rails).

Introduced low-level embedded search (Lucene and Whoosh) in APIs suitable for small data use cases, e.g. university search that factors in user preferences, academic strength and marketing requirements.

Departure of other Staff Engineers made me de facto lead on Platform Engineering, DevOps and Data Engineering as well as Backend and Architectural Lead.

Led a transition from a Spotify-style “chapter” model for disciplines to a “guild” model to encourage collective decision-making and empower engineers to shape architecture and technical choices.

Lead role across 2 teams distributed between Manchester, UK and Kirkland, USA (8 reports).

Built standard deployment platform on Kubernetes (AWS EKS), Quarkus, Terraform, GitHub Actions.

Mentored on Continuous Delivery, infrastructure as code and DevOps.

Built event-driven traffic alerting using Kafka, Elasticsearch (around 1 billion data points), RabbitMQ and Reactive Java (Quarkus).

Contributed to a company-wide AI project feeding data into LLMs.

Innovated Python project for applying BDD to an AI/LLM-based project using local LLMs for testing purposes.

Technical leader on highly-available Node.js and Spring Boot APIs used by whole BBC website (~50 internal clients, 15k requests per second for public-facing APIs).

Continued to work on Search technologies with broadening into analytics and micro-frontends.

Mentoring and leading on projects to promote knowledge sharing across up to 4 teams of engineers.

Drove architecture and data modelling for a unified approach to website analytics collection.

Developed lightweight feature flag mechanisms for previewing large-scale branding changes.

Promoted a unified CD (Jenkins) and deployment approach across 30+ services and 4 teams, supported with internal metrics collected in Elasticsearch.

Led on Python tooling for automating deployments, common infrastructure patterns and testing.

Primarily working on Search product with Solr, some bespoke APIs and some low-level Lucene.

Developed event-driven data pipelines with Java (Jersey), Apache Camel and ActiveMQ.

Built tooling in Ruby and Cucumber to drive projects via BDD and TDD.

Building data aggregation services in Java (Spring MVC)

Contributed to some front end in PHP and JS on the BBC Home page.

Creating geospatial BBC Weather search using Solr and PostGIS

Performance testing for a new Ethernet card and drivers on both Linux and Windows.

Created via bash scripts an early version of what today would be called Continuous Integration.

MSc (Distinction) in Software Engineering and Internet Architecture.

Awarded by Bradford, run in conjunction with the BBC.

Dissertation: Improving content discovery through combining linked data and data mining techniques

MA (Hons) (Cantab) in Computer Science.

Dissertation: Vector-based music search to generate playlists from qualitative descriptions