Chimnie is a British property data intelligence platform that brings deep, accurate, and transparent property insights to both consumers and businesses. It combines public, proprietary, and licensed datasets to build comprehensive property profiles covering valuations, risk factors, geospatial attributes, market insights, and more - for every residential and commercial property in the UK.

With over , Chimnie enables instant access to value and location insights for every property in Britain. By transforming complex, fragmented datasets into clear, accessible intelligence, Chimnie empowers homebuyers, property professionals, insurers and lenders to make faster, more confident decisions - whether researching a single address or analysing entire portfolios at scale.

Chimnie approached MajiLabs with a clear vision but no technical team in place. The goal was to that could validate the platform, attract early users, and lay the foundations for a long-term, scalable product.

The challenge was the . At the time, Chimnie needed to process and surface insights for , combining fragmented public and licensed datasets into a single, reliable source of truth. Delivering fast, accurate search and filtering across such a large dataset ruled out traditional database approaches and required a purpose-built search and indexing strategy using ElasticSearch.

Beyond data volume, the platform had to deliver instant results, intuitive filtering, and a consumer-friendly experience, while still meeting the accuracy and performance expectations of enterprise users such as insurers and property professionals.

ol All of this needed to be delivered at speed, with MajiLabs owning the entire technical delivery - from frontend and backend development to infrastructure, databases, AI and machine learning - and with the additional requirement that the system could later be handed over cleanly to an in-house engineering team.

To meet Chimnie’s need for speed without compromising long-term scalability, we adopted an backed by production-ready architecture from day one. The goal was to move fast, validate early, and avoid costly rewrites as the platform scaled.

We designed and delivered a Next.js application supported by a NestJS backend, deployed on Google Cloud Platform using Kubernetes. This provided the flexibility to iterate quickly while ensuring the system could handle increasing traffic and data volume. Infrastructure was fully managed using Terraform, enabling repeatable, auditable, and scalable deployments across environments.

Handling tens of millions of property records required careful data and performance considerations. An initial implementation using BigQuery proved too slow for the real-time filtering and instant search experience Chimnie required. We therefore re-architected the data layer around MongoDB paired with Elasticsearch, synchronising the two to deliver fast, reliable, and highly responsive search across the entire dataset.

Execution was delivered by a , covering frontend, backend, infrastructure, data engineering, and AI/ML. We worked in close collaboration with Chimnie through weekly check-ins, rapid iteration cycles, and Figma prototyping, ensuring alignment between product vision, user experience, and technical delivery throughout the build.