Blog
Thoughts on building, AI, genetics, and preventive healthcare
MaxToki: The AI That Predicts How Your Cells Age
A 1B parameter temporal model trained on 22 million age-annotated cells predicts how gene activity shifts across your lifespan.
AI Embedding Models Explained: Google, Perplexity & OpenAI Compared
Google released Gemini Embedding 2 and Perplexity released pplx-embed in the same week. Here's the science behind embeddings and how these models compare.
Protein Language Models: Treating Amino Acid Sequences Like Sentences
DNA is a language. Proteins are its stories. Here's how AI models trained on amino acid sequences are learning to read, write, and design biology from scratch.
KV Caching Explained: Why LLM Inference Is Memory-Bound
LLM inference isn't bottlenecked by compute. It's bottlenecked by memory. Here's how KV caching works, why it matters, and the math behind it.
The Math Behind LLMs
A single equation drives every large language model. Here's the math that keeps the world moving — explained without the hand-waving.
No posts found matching your criteria.