Hi! This is Prabhav. I am a Research Engineer in Meta SuperIntelligence team, and this is my notes page. The content is mostly around different aspect of machine learning research and engineering. Hope, you find it interesting!
Finished blog posts:
Understanding Latents: Variational Auto-Encoder
Modern machine learning models—from autoencoders to large language models—rely on the idea of latent representations.
KMeans Clustering
KMeans clustering is one of the most fundamental and widely-used unsupervised learning algorithms for partitioning data into distinct groups.
Research wikis
Interlinked notes I build up as I read papers — entities, concepts, and source summaries, cross-referenced.
- Audio LLM Wiki — Moshi, Voxtral, Qwen3-Omni, and Thinking Machines’ interaction models.