Parinaya Chaturvedi

This is the website of Parinaya Chaturvedi. I work on AI, NLP, and LLMs. If you like these posts, subscribe to the mailing list.

• vLLM PagedAttention: Efficient Memory Management for LLM Inference

  December 10, 2024

  Understanding PagedAttention, the memory optimization technique that enables efficient serving of large language models.

  ---

• Sample Blog Post

  January 15, 2025

  This is a sample blog post to demonstrate the website functionality.

  ---

• Pipeline-Parallelism: Distributed Training via Model Partitioning

  October 03, 2022

  Pipeline parallelism makes it possible to train large models that don't fit into a single GPU's memory.

  ---

• GPU CUDA Primitives: Understanding Parallel Computing Fundamentals

  November 15, 2024

  A deep dive into CUDA primitives, memory hierarchies, and parallel computation patterns on modern GPUs.

  ---