Case studies

Streamline auditing assessments for UK tax services (2023 — 2024)

Brief:
(1) Improve the quality and design of auditing data across HMRC and (2) automate the assessment process as much as possible

Solution:
Matthew created comprehensive new auditing guidance for UK tax services, reducing reliance on human assessors, asynchronous communication and meetings to enforce quality. For the first time, it was written into a clear set of internally published guidelines. Matthew went on to conceive an innovative tool for teams to check their proposed audit events prior to a formal assessment. Matthew creating the wireframes and led development of the tool which went live in January 2024

Technologies:
Scala, Play Framework

Impact:
A streamlined and more consistent audit assessment process for tax services with less reliance on humans. Investigators and data analysts can more easily query and work with auditing data.

Audit quality self-service tool for third party software vendors (2022 — 2023)

Client:
Equal Experts and HMRC

Brief:
Help the team launch the product successfully and support it afterward

Solution:
Matthew helped the team successfully launch a new product that enabled vendors of accounting software that communicate with HMRC to check their auditing data. Matthew accelerated the launch using his prior extensive experience working with the HMRC MDTP tax platform, full-stack engineering, AWS experience and techniques to satisfy the WCAG 2.1 AA accessibility standard.

Following launch, Matthew helped support, maintain, and improve the performance of the product in a live-running phase for the next 12 months. Improvements included resolving longstanding performance issues in two critical data pipeline components, achieving a 10x reduction in disk space usage for a Kafka Streams application and a 9x improvement in the running time of a Scala Apache Spark application.

Technologies:
Scala, Play Framework, Opensearch, Apache Spark, Airflow, Kafka Streams, Cats, AWS, PostgreSQL, Terraform

Impact:
The team were upskilled and motivated by successful delivery that maintained and enhanced their reputation. An improved developer experience for third party software vendors.

Engineering management consultancy for a multinational news and information services company (2022)

Client:
Axiologik and Thomson Reuters

Brief:
(1) Design an automated self-assessment tool for product engineering teams and (2) advise management on ways to reconcile requirements of traditional service management with DevOps and agile software development

Solution:
Matthew designed and subsequently oversaw the development of an automated self-assessment tool for product engineering teams that enabled them to identify quality gaps and improve their engineering practices as well as help management target significant technology investment. Matthew advised directors and senior management on ways to introduce effective governance without slowing teams down with bureaucracy through integrating agile software development tools with traditional service management platforms. Matthew designed a risk assessment algorithm and tool to route changes along the most appropriate path through service management, applying the right level of scrutiny according to the safety of the change.

Technologies:
Microsoft Forms, Microsoft Power Automate, PowerBI

Impact:
The tool was successfully rolled out to all of Thomson Reuter’s highest grossing enterprise products and helped inform significant investment decisions. Senior management were influenced and helped to reconcile traditional service management with agile product delivery

Scala engineering consultancy for an energy technology company (2021 — 2022)

Client:
Kaluza Ltd (an Ovo group company)

Brief:
Help the team launch and scale a new energy charging pipeline successfully

Solution:
Matthew accelerated the launch of a new energy charging pipeline utilising a complex tech stack and serving millions of households in the UK. Matthew led work to scale the pipeline to handle an additional 10x supplies by introducing an event-sourcing technique known as snapshotting achieving a 78% reduction in processing time, introduced a circuit breaker to monitor and react to data volatility and made major contributions to the deployment of the pipeline to Australia.

Technologies:
Pure functional Scala, Cats Effect, Kafka, PostgreSQL and AWS EKS (Kubernetes)

Impact:
An enhanced reputation for the team and elimination of lag in the energy pipeline resulted in fresher data for downstream components including billing

Python engineering consultancy for an AI startup (2021)

Client:
kaizan.ai

Brief:
Build out a data processing pipeline

Solution:
As one of the first engineers at this innovative AI start-up, Matthew helped the team build out a data processing pipeline that formed the backbone of this AI product. He collaborated closely with the founder and ML engineer to integrate their work into the platform and maintained a high standard of code quality through a test-driven approach.

Technologies:
Python, Django, Celery, PostgreSQL, NLTK, OpenAI

Impact:
Matthew's contributions were instrumental in shaping the company’s early success and establishing a solid technical framework for future growth.

Scala and Typescript engineering consultancy for the UK tax collector (2020 — 2021)

Client: Equal Experts and HMRC

Project 1

Brief:
Deliver a cookie consent solution for HMRC

Solution:
Matthew led the development of a centralized cookie consent solution that allowed tax users to choose whether to store non-essential cookies in their browser. The delivery was particularly challenging because it had to work seamlessly across hundreds of independently built frontend microservices, rely on each team making a small change to their codebase, and scale to handle tens of thousands of requests per minute.

Technologies:
Typescript, Scala, Play Framework

Impact:
The project was delivered successfully amid the Covid-19 pandemic, delivering enhanced privacy for the public and reducing compliance risks for the UK tax collector.

Project 2

Brief:
Deliver a solution to ensure HMRC meet its responsibilities under UK accessibility legislation

Solution:
Matthew scoped and led the development of a service to allow teams to create and publish standardised accessibility statements. The service allowed HMRC to successfully meet a pressing hard regulatory deadline in September 2020.

Technologies:
Scala, Play Framework, Circe

Impact:
The project was delivered on time during the Covid-19 pandemic, ensuring HMRC met its statutory accessibility obligations and improved compliance management across its digital estate, enhancing access for users with disabilities.

Technical leadership for a leading price comparison website (2019)

Client:
uSwitch.com and money.co.uk

Brief:
Help a team unfamiliar with a chosen technology stack deliver an ambitious re-platforming project

Solution:
Matthew provided technical leadership and guidance for a team new to the technology stack, overseeing the migration of the financial and insurance sections of the uSwitch and money.co.uk websites to a new platform. He improved communication and collaboration across various teams, introduced unit testing, static code analysis, and pair/mob programming, all while maintaining a positive team atmosphere.

Technologies:
React, GraphQL, Node.js, Clojure, Kubernetes

Impact:
The project was successfully delivered, resulting in a streamlined technology stack for the organization

Smart in-home technology asset management solution (2018 — 2019)

Client:
Ovo Energy Ltd

Brief:
Create a database and asset management system for smart meters to replace a spreadsheet-based process

Solution:
Matthew architected and developed a comprehensive solution to manage the lifecycle of millions of in-home technology assets, including smart meters

Technologies:
Node.js, Flow (a statically typed language very similar to Typescript), AWS lambda, AWS Fargate, containers, PostgreSQL, DynamoDB, SQS, Kafka, Rest APIs, React, Ant Design, Terraform, CI/CD, CircleCI, TDD/BDD, Mocha, Sinon, Chai, Enzyme, Cucumber and Webdriver IO

Impact:
The new system eliminated reliance on spreadsheets, establishing a robust relational database as a single source of truth

Smart meter installer mobile app (2018 — 2019)

Client:
Ovo Energy Ltd

Solution:
Matthew designed and built the back-end infrastructure for a React Native mobile app used by SMETS2 smart meter installation engineers nationwide

Technologies:
Terraform, Packer, AWS ECS/Fargate, Spring Boot, Kotlin, CircleCI, PostgreSQL

Impact:
The smart meter installer app was rolled out successfully and was being successfully used in production by the end of the engagement