Multimodal Deepfake Detection
Why image-only detection is not enough—and how image, video, and audio pipelines can work together to assess the integrity of modern media.
I write about the parts of technology that become clearer only after building, breaking, debugging, and questioning them. My articles cover deepfake detection, computer vision, data engineering, geospatial systems, and the lessons hidden inside technical failures.
Why image-only detection is not enough—and how image, video, and audio pipelines can work together to assess the integrity of modern media.
A practical account of developing a multi-face recognition system, from face detection and classification to backend architecture and real-time performance.
A reflection on debugging as more than error correction: a method for learning how systems behave, testing assumptions, and becoming a stronger engineer.
A comparison of two approaches to organising and querying geospatial data, with practical trade-offs for performance, maintainability, and production workflows.