Core member of the R&D team at SynergyLabs — implementing SOTA deep learning research for production deployment, building edge-ready computer vision systems, and representing the company at AI conferences and in published research.
QC Tool with Cobot Integration
Developed a suite of image preprocessing techniques capable of performing automated quality control tasks, integrated with a Universal Robots cobot for physical inspection pipelines. The system combined computer vision defect detection with robotic actuation — enabling real-time QC decisions on the production line without human intervention.
R&D & SOTA Research Implementation
As a member of the core R&D team, implemented a range of state-of-the-art research papers for production use — including DenseCap (dense image captioning) and FaceNet (facial recognition). The facial recognition attendance system achieved 97.8% accuracy in real-world deployment conditions.
Work spanned the full research-to-production cycle: reproducing paper results, adapting architectures for edge deployment constraints, and building inference pipelines optimised for latency.
AI Conference Representation & Published Research
Represented the company's corporate development division at several AI Summits, presenting applied deep learning work to industry and academic audiences. Recognised at the Global AI Summit 2019. Also published research in machine learning and deep learning — contributing to the broader academic record in computer vision and edge AI.
