I am a research scientist at Pinecone where I lead the investigation of efficient and effective algorithms and probabilistic data structures for constrained Maximum Inner Product Search over dynamic, heterogeneous vectors, as well as learning to rank in vector spaces. My research interests also include the theory of stochastic ranking policies and a multifaceted study of efficiency, from time and space complexity to resource and energy consumption.
I co-organized the first international workshop on Reaching Efficiency in Neural Information Retrieval (ReNeuIR) at SIGIR 2021 with Claudio Lucchese and Franco Maria Nardini. (Workshop Website, Article)