https://thegustafson.com https://thegustafson.com/series https://thegustafson.com/map https://thegustafson.com/toybox https://thegustafson.com/blog https://thegustafson.com/about https://thegustafson.com/blog/the-missing-middle https://thegustafson.com/blog/vectors-and-spaces https://thegustafson.com/blog/norms-dots-similarity https://thegustafson.com/blog/probability-distributions https://thegustafson.com/blog/cross-entropy-and-loss https://thegustafson.com/blog/gradients-and-learning https://thegustafson.com/blog/optimizers https://thegustafson.com/blog/gpus-and-floating-point https://thegustafson.com/blog/prehistory-of-nlp https://thegustafson.com/blog/next-token-prediction https://thegustafson.com/blog/ngrams-and-sparsity https://thegustafson.com/blog/word2vec https://thegustafson.com/blog/glove-fasttext https://thegustafson.com/blog/rnns https://thegustafson.com/blog/lstms-grus https://thegustafson.com/blog/seq2seq-and-bahdanau https://thegustafson.com/blog/unicode-and-bytes https://thegustafson.com/blog/bpe-from-scratch https://thegustafson.com/blog/wordpiece-unigram-sentencepiece https://thegustafson.com/blog/vocabulary-design https://thegustafson.com/blog/embedding-table https://thegustafson.com/blog/special-tokens-chat-templates https://thegustafson.com/blog/packing-and-masking https://thegustafson.com/blog/attention-from-scratch https://thegustafson.com/blog/multi-head-attention https://thegustafson.com/blog/causal-masking https://thegustafson.com/blog/positional-encodings-and-rope https://thegustafson.com/blog/ffn-as-memory https://thegustafson.com/blog/layer-norms https://thegustafson.com/blog/decoder-vs-encoder-decoder https://thegustafson.com/blog/reading-attention-is-all-you-need https://thegustafson.com/blog/moe-layers https://thegustafson.com/blog/training-vs-inference https://thegustafson.com/blog/prefill-vs-decode https://thegustafson.com/blog/the-one-new-row https://thegustafson.com/blog/kv-cache-from-first-principles https://thegustafson.com/blog/sampling-strategies https://thegustafson.com/blog/speculative-decoding https://thegustafson.com/blog/continuous-batching https://thegustafson.com/blog/prefix-caching https://thegustafson.com/blog/structured-generation https://thegustafson.com/blog/what-an-inference-engine-does https://thegustafson.com/blog/kernel-fusion https://thegustafson.com/blog/paged-attention https://thegustafson.com/blog/memory-management https://thegustafson.com/blog/quantization https://thegustafson.com/blog/parallelism-strategies https://thegustafson.com/blog/throughput-vs-latency https://thegustafson.com/blog/comparing-engines https://thegustafson.com/blog/benchmarking-inference https://thegustafson.com/blog/pretraining-data https://thegustafson.com/blog/distributed-training https://thegustafson.com/blog/sft https://thegustafson.com/blog/rlhf https://thegustafson.com/blog/dpo-and-variants https://thegustafson.com/blog/constitutional-ai https://thegustafson.com/blog/tool-use-finetuning https://thegustafson.com/blog/long-context-training https://thegustafson.com/blog/synthetic-data-distillation https://thegustafson.com/blog/lora-and-peft https://thegustafson.com/blog/loss-vs-benchmarks https://thegustafson.com/blog/eval-harnesses https://thegustafson.com/blog/contamination https://thegustafson.com/blog/evaluating-reasoning https://thegustafson.com/blog/tool-agent-evals https://thegustafson.com/blog/human-preference-eval https://thegustafson.com/blog/calibration-and-abstention https://thegustafson.com/blog/embeddings-and-vector-search https://thegustafson.com/blog/vector-search-and-ann https://thegustafson.com/blog/dense-sparse-hybrid https://thegustafson.com/blog/chunking-strategies https://thegustafson.com/blog/rerankers https://thegustafson.com/blog/rag-end-to-end https://thegustafson.com/blog/graphrag https://thegustafson.com/blog/context-windows https://thegustafson.com/blog/context-engineering https://thegustafson.com/blog/history-of-agents https://thegustafson.com/blog/function-calling https://thegustafson.com/blog/the-agent-loop https://thegustafson.com/blog/transcript-formats https://thegustafson.com/blog/mcp https://thegustafson.com/blog/remote-mcp-and-auth https://thegustafson.com/blog/planning-and-reasoning https://thegustafson.com/blog/computer-use https://thegustafson.com/blog/multi-agent https://thegustafson.com/blog/openai-api https://thegustafson.com/blog/anthropic-api https://thegustafson.com/blog/google-api https://thegustafson.com/blog/open-weight-models https://thegustafson.com/blog/huggingface-ecosystem https://thegustafson.com/blog/cloud-model-marketplaces https://thegustafson.com/blog/lock-in-and-portability https://thegustafson.com/blog/scaling-laws https://thegustafson.com/blog/inference-economics https://thegustafson.com/blog/open-vs-closed https://thegustafson.com/blog/safety-and-alignment https://thegustafson.com/blog/ai-regulation https://thegustafson.com/blog/standards-wars https://thegustafson.com/blog/history-2013-2026 https://thegustafson.com/blog/what-comes-next