Life is not a race, but a journey

Notes and summaries for CMU 15-645 Database Systems.

Notes on NLP evaluation, benchmarks, and experimental design from CMU 11-711 Advanced NLP, covering loss-based metrics, task metrics, LLM-as-judge, statistical testing, and annotation quality.

Notes on multimodal modeling from CMU 11-711 Advanced NLP, covering ViT, CLIP, LLaVA-style integration, token-based image generation, diffusion, and flow matching.

Notes on retrieval systems, retrievers, and retrieval-augmented generation (RAG) from CMU 11-711 Advanced NLP.

Notes and summaries for CMU 15-642 Machine Learning Systems on distributed training techniques including data parallelism, model parallelism, pipeline parallelism, and optimization strategies like ZeRO for training large-scale models.

Notes on decoding strategies for LLM inference from CMU 11-868, covering efficient sampling, beam search algorithms, speculative decoding, and EAGLE.

Comprehensive notes on tokenization techniques and vocabulary learning for Large Language Models, covering BPE, VOLT, and practical considerations in multilingual LLMs.

Notes and summaries for CMU 15-642 Machine Learning Systems on Transformer architecture, attention mechanisms, and GPU optimizations including FlashAttention.

Notes and summaries for CMU 15-645 Database Systems.

Implementation and architecture of TinyKV, a Raft-based distributed key-value storage system covering Multi-Raft, MVCC transactions, and the scheduler.