Trained on a vast dataset of furniture styles, décor, textures and materials — the AI intelligently suggests design elements, arranges furniture to complement existing layouts while accounting for shadow casting, natural light adjustments and perspective alignment. Users visualise different options and swap furniture in real-time.

Advanced object detection and tracking algorithms for real-time identification of suspicious activities. Rigorous testing and iterative model fine-tuning delivered a robust system reliable even in complex multi-person environments with occlusions and varying distances.

Trained on a diverse dataset of waste items, the system accurately identifies and sorts recyclables, compost and landfill materials — significantly improving sorting precision, reducing contamination and promoting efficient waste processing pipelines.

Custom AI model for virtual garment try-on directly on user photos — advanced computer vision and deep learning for realistic, accurate fitting. Dramatically improves user confidence and engagement in online fashion shopping. Live at fashion360.com.au.

Advanced image recognition identifies ingredients from photos — generating creative, tailored recipes based on what's available. Iterative refinements achieved high accuracy even under challenging real-world capture conditions. Open source on GitHub.

Cutting-edge AI for seamlessly integrating faces into artistic masterpieces — fine-tuned for precision and realism across painting styles, historical portraits and contemporary art. Published as a Chrome Extension on the Web Store.

Trained on a comprehensive dataset covering various plant types and health scenarios — highly accurate diagnoses and personalized care recommendations empower users to effectively nurture their plants based on a single photo submission.

Developed in close collaboration with real graffiti artists — fine-tuned to capture the authenticity, creativity and dynamic style of street art. Transforms any photo into a striking, graffiti-inspired masterpiece with true artistic fidelity. Listed on There's An AI For That.

Robust AI for precise sentiment analysis in dense, high-complexity scenes — computer vision and deep learning handle occlusion, distance variation and overlapping individuals to deliver accurate emotional intelligence from live environments.

Fine-tuned AI recommendation system that seamlessly connects users with adoptable pets matching their criteria — integrates shelter databases across Ireland, understands breed traits and lifestyle preferences to enhance pet adoption outcomes. Live at petmatch.ie.

AI trained on a diverse dataset of deteriorated images — intelligently repairs and revitalises damaged photos, recovering precious memories with remarkable detail and clarity. Published on the Apple App Store with strong user ratings.

AI that generates engaging, perfectly balanced spot-the-difference puzzles — iterative refinement ensures each puzzle offers the ideal level of challenge while maintaining visual appeal. Live production game at spotthedifference.com.

Dynamic AI system that adapts to users' interests and real-time location — providing tailored travel recommendations and engaging visual content. Personalises the sightseeing experience based on preferences, history and geolocation. Live at ai-adventures.com.

Trained on a wide range of artistic techniques — transforms photos into unique, stylised sketches while maintaining the original character and personality of each image. Live at sketchme.ai with thousands of daily active users.

Robust AI system for analyzing complex infrastructure and environmental data from drone footage — accurate real-time assessments and actionable intelligence for construction monitoring, maintenance and environmental impact analysis.

Curated and annotated a comprehensive dataset covering multiple sign languages — generates accurate video translations for the deaf and mute community, interpreting diverse gestures reliably. Live at slait.ai.

Advanced AI generating accurate detailed 3D face models from a single photo — adapts to lighting and expressions, suitable for virtual environments, identity verification and gaming. Live at facegen.com.

AI system that recognises and finds visually similar clothing items from social media photos across online stores — advanced image recognition and ML deliver a seamless fashion shopping discovery experience.

Secure AI for reliable real-time license plate recognition — empowers law enforcement with monitoring tools while preserving data privacy. Robust across varying lighting, angles and weather conditions.

Computer vision + recommendation engine powering interactive smart fitting rooms — users virtually try outfits and receive customised style suggestions based on preferences, body shape and behaviour, delivering a seamless in-store AI experience.

Predictive model forecasting product sales with ~87% accuracy — enabling informed decisions for inventory management and marketing strategy. Built with Python and Keras in 2 months.

Robust system continuously monitoring and analyzing corporate network traffic — promptly alerts on suspicious or anomalous activity, enhancing security and risk management across enterprise environments. 4 months.

ML solution that accurately replaces text on any surface in images — preserving original font style, size and indentation. The change is visually indistinguishable from the source. Python and PyTorch, 1.5 months.

Simulation environment and detailed robotic hand model — trained ML models to master different mini-games, demonstrating adaptive precision and learning capabilities across diverse gaming scenarios. 3 months.

4-camera AI motion tracking — tracks human movements for seamless real-time interaction, delivering personalised information based on gestures. Unity visuals + Python AI + React.js/Node.js backend. 6 months.

Users generate unique AI artwork and mint NFTs on blockchain — neural networks process text prompts into digital assets, Solidity smart contracts ensure secure ownership. React frontend, Node.js backend. 2 months.

AI + computer vision for 96% accuracy aircraft tracking from a static camera — reliable under varying environmental conditions. Deployable at airports and air traffic monitoring stations. OpenCV, 1 month.

Semantic parsing tool that breaks down sentences and generates sign language video translations — making text content accessible to individuals with hearing and speech impairments. Real-time processing, 1 month.