Drax Energy is a UK-based company that provides renewable source electricity, electrification services and smart meters to businesses and organisations,

I worked with their tech and delivery teams to ideate, discover and implement Hyperautomation AI projects across the organisation.

Background & Objectives


The Accounts Payable process for Drax UK Generation showed approx. 107 hrs manual effort per month that could benefit from automation efficiency and risk avoidance. An early pilot of key analysis into the process showed up to 70% of the team’s time was taken up on two manual tasks.

Multiple escalations (250 per week, 50% of these derive from front-end processing) are raised regularly due to human error, lack of balance on supplier purchase orders and poor OCR quality.

There are also opportunities to scale and extend to other business areas and teams (UK and Globally).


The current process involves significant manual effort, multiple escalations due to human error and lack of balance on supplier purchase orders, and wasted time due to email and OCR handling.

The key objectives were:

  • Review and endorse one of two high-level solution options to automate the UK Generation Accounts Payable invoice inbox processing.
  • Deliver ROI within 12 months
  • Reduce error rate through higher quality ML models
  • Process data for accurate ESG reporting


As the AI Principal and Enterprise Architect on the Drax Hyperautomation project, my responsibilities were:


Architect enterprise platform solutions for integrating with multiple data sources, third-party providers and advanced computer vision ML models and SaaS providers

Stakeholder Management

Work with stakeholders across the management team to agree on project requirements, build the enterprise architecture, benefits model, ROI model and project delivery plan


As an SME, responsible for the strategic direction of the AI tools, solutions and governance to dictate the best approach to achieving business outcomes

Security & Privacy Governance

Discussion and ideation of new architecture and security governance framework in line with qualified use cases

Enterprise Integrations

Third-party enterprise technical integrations with multiple providers including IBM Maximo, Infor, Rossum, Azure, Solace, Snaplogic


Enterprise Platform Solution

We successfully delivered a end-to-end enterprise platform solution that reduced the manual processing time from 110 hours per month to ~5 hours; a 95% improvement to be delivered within 3 months. We build an end-to-end data pipeline, accelerated with machine learning, that automated the AP journey. It validated incoming email documents, extracted document entities and updated their ERP systems for a human to approve at the end.

Additionally due to the processes implemented, we saw 4-5 times more flight opportunities by scaling the hyperautomation team with the right skills and roles and qualified all incoming opportunities and prioritized accepted ones onto a structured backlog.

We also identified and prioritized automation recommendations on IT/business area backlogs and delivered at least one IDP solution to production with a positive ROI, where forecast cost savings or benefits were greater than the implementation and operational costs.

Tools & Technologies

This project branched into many varying aspects of engineering and technologies, and a few tools were used.

  • Architecture
  • Swagger
  • Jira
  • Confluence
  • AI
  • Data
  • AWS
  • Azure
  • GCP
  • Tensorflow
  • Microsoft Office Suite