Three Mission-Critical Platforms for UK Energy Transition Leader

energy platform development case study

Client

The client is a UK-based energy company leading the transition to a low-carbon future. It provides services and technology that help businesses and households reduce their carbon footprint and manage energy consumption more efficiently, across the residential and commercial market.

About

Industry

Energy, low-carbon transition

Company size

50-100

Client Market

Global

Location

UK

Challenge

Specialised skills needed across three parallel builds

The client needed to build three technically demanding energy platforms at once but lacked the specialised in-house expertise to deliver any of them alone.

Building energy grid software calls for a rare combination of expertise: real-time systems, energy market regulation, enterprise integration at scale, and cloud architecture suited to data-intensive platforms. The client recognised it did not have this specific expertise internally.

Hiring senior engineers with this profile would take six months or more in the competitive UK tech market, and building a full platform engineering team from scratch would mean significant additional investment in recruitment and onboarding. With three critical platforms needing simultaneous development to meet market timing, the client needed a partner able to bring architecture expertise and hands-on delivery straight away.

The client’s team understood its market and what it needed to build, but required a partner able to bring grid engineering, market systems and enterprise integration expertise together immediately, across three platforms at once.

Solution

Three platforms,
one embedded team

A flexible grid management system for distributed energy resources

Built to coordinate thousands of distributed energy resources including solar panels, batteries and EV chargers in real time, using a dual-database architecture and asynchronous processing to keep response times within grid stability limits.

A market-wide half-hourly settlement system

Delivered as one of three platforms built in parallel with shared cloud infrastructure and common engineering patterns on AWS, so security and deployment practices proven on one platform could be applied rapidly to the others.

An enterprise integration platform connecting field operations to Dynamics 365

Architected for real-time, bidirectional data exchange rather than batch processing, so field engineers installing EV chargers and smart meters get immediate access to customer and service data, with installation data flowing back for accurate billing straight away.

A 10+ person cross-functional Zartis engineering team was embedded across all three platforms throughout delivery.

The Zartis Approach

What we delivered

Time-series architecture for grid telemetry

Grid telemetry (voltage, frequency, power flow) behaves fundamentally differently from configuration or user data, generating high volumes of time-stamped sensor readings that a conventional relational database is not built to handle efficiently.

Challenge: Storing high-volume grid sensor data in a traditional relational database created performance bottlenecks and storage inefficiencies.

Solution: A dual-database approach paired an open-source time series database (TSDB) with a separate database for relational data.

Asynchronous processing for real-time coordination

Grid operators need to adjust power output across thousands of devices simultaneously, for example when responding to frequency drops, and synchronous processing would introduce unacceptable latency into that response.

Challenge: Coordinating commands and confirmations across thousands of distributed devices in sequence was too slow for grid conditions.

Solution: The selected Python framework handles thousands of concurrent device connections in parallel.

Security and availability by design

Live grid infrastructure carries higher security and reliability stakes than most software platforms; a bug or breach here can destabilise electricity supply for thousands of homes and businesses.

Challenge: Standard platform security and availability standards were not sufficient for infrastructure controlling live grid operations.

Solution: A zero-trust model with circuit breakers and safe fallback behaviours was built into the platform from the outset.

Open protocol for device interoperability

A proprietary communication protocol would lock the client into specific hardware vendors, which works against onboarding new device types and manufacturers as the market grows.

Challenge: Proprietary device protocols would restrict which hardware the platform could integrate with over time.

Solution: The OpenLeadr (Open Automated Demand Response) international standard was implemented for device communication.

Real-time integration with Dynamics 365

Field engineers need immediate access to customer and service data on site, and batch processing would delay both installation work and the billing data flowing back from it.

Challenge: Batch-based integration would create delays affecting installation efficiency and billing accuracy.

Solution: A bidirectional, real-time integration was built between field operations and Microsoft Dynamics 365.

Business Impact

Coordinating distributed energy resources in real time

Maintaining grid stability through regional cloud disruption

Connecting field operations directly to core billing systems

Faster onboarding of new energy devices and vendors

Supporting the shift toward renewable-led grids

Results

Built for live grid conditions

100% uptime

During a major regional cloud service disruption, the platform maintained full uptime through automatic failover, validating the multi-region architecture built to protect live grid infrastructure.

<1s response times

The asynchronous device-coordination architecture enables grid operators to adjust output across thousands of distributed energy resources within the UK’s regulatory frequency range of 49.5-50.5 Hz.

10-100x faster queries

The dual-database architecture delivers time-series query performance an order of magnitude faster than a single relational database, supporting real-time grid optimisation.

Slide 1
Case Studies
zartis case study slider active
zartis case study slider element
zartis case study slider element
zartis case study slider element
zartis case study slider element
It's time to
reach further