
Quadtrees for Space Decomposition, Java Implementation
August 18, 2012The quadtree structure will be utilised for the geometric clustering, by recursive decomposition, of the locations of graph vertices on a 2D plane.
I help startup founders build and launch products with a lean, founder-focused approach.
I'm Kirsty, a senior software engineer with a background in both design and computer science. Over the last 15 years, I've worked with early-stage startups in the US, UK, and Canada β helping founders turn ideas into scalable products. With experience in both San Francisco and the UK, I thrive in small, fast-moving teams, especially at the seed and Series A stage. I wear many hats β engineering, product, UX β and bring a Swiss Army knife skill set to every project. I'm passionate about building intuitive, high-quality experiences that balance design and technology. Alongside my solo work, I run StackTiger, a micro-agency where we build MVPs for startups.
Designed, built, and launched ChainLobby, a feature-rich Web3 app on Cardano using React - optimizing SQL queries, designing recommendation algorithms, scaling infrastructure for millions of blockchain transactions, implementing NFT image processing, and streamlining deployments with CI/CD.
ChainLobby had typical social network features, group chat features with custom slash commands, a wide array of community customisation features, token-gated content, and rich article editing capabilities.
Built and scaled a React Native college chat app with 250K+ active users, optimizing performance, feature development, and a high-volume chatbot.
As primarily a member of the backend team, key contributions included building out the chatbot, implementing a custom push notification system, and optimizing the backend for high-volume chat.
I also worked on the frontend, implementing new features, optimizing performance, and fixing bugs.
The app was built with a React Native frontend targeting iOS and Android, a Node.js backend, and a SQL database, and was deployed using CI/CD pipelines.
Designed and developed the Daily Stat Cards React Native app, integrating Firebase push notifications to deliver real-time football statistics and personalized alerts, helping users stay informed and make data-driven decisions.
Built an interactive document and report tool for healthcare data, allowing users to structure reports with shortcuts, drag-and-drop, and real-time data annotations. Enabled seamless transitions between editing and a published interactive mode, giving readers the ability to explore and interact with healthcare insights in-place.
I was awarded an EPSRC scholarship for research at the University of Liverpool, studying time and memory-efficient methods for performing analysis of large-scale networks.
The main emphasis was on efficient visual presentation of structural properties and significance of certain components of graphs.
I developed a number of algorithms and UI tools using Java.
The quadtree structure will be utilised for the geometric clustering, by recursive decomposition, of the locations of graph vertices on a 2D plane.