CS PhD from Stanford NLP with previous experience at Meta and Princeton CS
Referenced in coverage
“Announced Fast Byte Latent Transformer accepted to ICML 2026, using diffusion for parallel byte decoding to make byte-level language models less inference-bound.”