EVEEVO

Introduction

EVEEVO is a UK-based digital marketplace focused on electric and hybrid vehicles (EVs). The platform is designed to improve how consumers discover, evaluate and purchase EVs by combining vehicle listings with richer, data-driven insights.

The business was founded to address a growing gap in the market. As demand for EVs increase, many consumers lack access to clear and relevant information to support decision-making. EVEEVO aims to bridge this gap by helping users understand how a vehicle fits their lifestyle before making a purchase.

The challenge: supporting better decisions in a complex EV market

The transition to EVs has introduced additional complexity into the car-buying process. Consumers must consider factors such as real-world battery range, charging behaviour and day-to-day usability, yet often struggle to access clear, tailored information.

Dealerships also face operational challenges. Many are managing high volumes of repetitive enquiries, while also missing opportunities outside of working hours. This can lead to inefficiencies, with time spent on unqualified leads and limited capacity to respond quickly to customer demand.

EVEEVO set out to address these challenges by developing an AI-powered sales agent capable of answering detailed, context-specific questions and supporting users throughout their decision-making journey.

The long-term ambition was to create a fully integrated AI sales agent. Recognising that this would require significant technical capability and investment, the focus shifted to building a strong foundation for future growth. Through the project, the team developed the core engine and a working proof of concept, validating the technology and demonstrating its commercial potential. This milestone positioned the company to seek additional investment and accelerate progress towards its wider vision.

The approach: developing a scalable AI foundation

The National Innovation Centre for Data (NICD) delivers activity through the Hartree Centre North East Hub as part of the wider Hartree Centre SME Hubs, which provide regional advanced digital technology support to UK industry. Through the Innovate UK BridgeAI programme, EVEEVO partnered with NICD to design and develop the core AI architecture underpinning the platform’s AI sales agent.

The project focused on creating a proof-of-concept chatbot capable of answering EV-specific customer queries using EVEEVO’s proprietary vehicle data.

Following initial research into retrieval augmented generation (RAG) approaches for structured vehicle data, NICD identified that a JSON-based approach would provide more effective and scalable results than a conventional vector embedding pipeline for constantly updating dealership data.

To deliver this, the team developed a series of JSON parsing agents capable of exploring and responding to queries across different vehicle datasets. These agents were integrated into a supervisor agent architecture alongside a conventional RAG vector store for static reference documents.

NICD provided technical guidance throughout, helping EVEEVO understand how to structure the system to support long-term growth.

"That's where you need advice from these teams... to determine the core architecture you should build to make it all work.”

Rebecca Jackson, CEO, EVEEVO

The project was delivered through a collaborative process, with regular engagement between teams, iterative development, and weekly progress updates. A working codebase was provided at the end of the project, enabling EVEEVO to continue development and integration.

The solution: a core AI sales agent

The project delivered a working proof-of-concept AI chatbot system tailored to EVEEVO’s use case.

The system was designed to answer detailed queries relating to electric vehicles, including information about mileage, battery capacity, range, dimensions and other model-specific features through a single natural language interface.

By carefully designing the tools and prompts used within the system, NICD ensured that responses remained grounded in verified vehicle data while reducing the risk of hallucinations from the large language model.

Alongside the chatbot itself, NICD also delivered:

a demonstration graphical user interface
the underlying source code for future integration
testing and evaluation of system responses against verified database answers

This provided EVEEVO with a technically validated proof-of-concept that can now be integrated into the company’s wider platform development roadmap.

Outcomes and impact: enabling growth and innovation

The project has provided EVEEVO with a strong technical foundation to support ongoing product development. With the core system in place, the business is now able to build additional features and move towards commercialisation.

Rebecca Jackson, CEO of EVEEVO adds:

“Now we’ve got the code… we can build on that and commercialise it.”

For dealerships, the solution has the potential to improve efficiency by handling routine enquiries and enabling engagement outside of standard working hours. This can reduce missed opportunities and allow teams to focus on higher value interactions with customers who are further along in their decision-making process.

For consumers, the platform supports more informed decisions by enabling users to explore detailed information before contacting a dealer. This helps reduce uncertainty and improves confidence when selecting a vehicle.

At a wider level, the solution has strong potential to scale across the automotive sector. By improving both customer experience and operational efficiency, it supports a more effective and data-driven approach to EV sales.

Looking ahead

With the core AI engine now in place, EVEEVO is well positioned to expand its platform and introduce additional functionality.

Future development will focus on extending the capabilities of the AI assistant, integrating with external platforms, and scaling the solution across the automotive sector.

“From a technical perspective this was an interesting project, grounding an LLM not just in text documents as I’d done before but using EVEEVO’s large electric car sales database. It was great to see how well the tools I’d written and the final multi agent system worked after just a few weeks of development, and I could really see the value the Bridge AI scheme was enabling us to deliver.”

Dr Chris Wedge, Data Scientist, National Innovation Centre for Data