Ed Pizzi

I'm an applied AI researcher and software engineer whose work focuses on representation learning and large-scale retrieval problems, including copy detection.

I'm the lead researcher for SSCD (CVPR 2022), a copy detection fingerprint that is now widely used for deduplicating foundation model training datasets (Llama 3, Stable Diffusion 3, DINOv2, DINOv3, MovieGen).

My recent work can be found under research. My earlier work spans several software engineering fields.

You can find me on Google Scholar, LinkedIn, Github.