How embeddings work
From tokens to vectors: how embedding models build semantic space, why similarity metrics work, and how dimensionality and model choice shape retrieval.
EmbeddingsVector DatabaseRAG
Research
AI research explainers
Deep, conceptual explainers on how AI systems actually work — vector databases, embeddings, transformer architecture, inference optimization, and evaluation. Written to explain the machinery, not just name it.
LLM evaluation methods
A taxonomy of LLM evaluation: benchmarks, LLM-as-judge, human eval, and task-specific metrics — what each measures and when it applies.
LLM EvalsAI EngineeringTesting
LLM inference optimization
How LLM serving actually works: KV cache, continuous batching, quantization, and speculative decoding — and the throughput-versus-latency tradeoff.
LLM InferencevLLMProduction AI
Transformer architecture explained
Attention, positional encoding, and stacked layers: how the transformer works and why it scales, for engineers who build on top of LLMs.
TransformersLLM ArchitectureAI Engineering
What is a vector database?
How vector databases work: ANN indexes (HNSW, IVF, PQ), the recall–latency–memory tradeoff, and when you need one versus pgvector.
Vector DatabaseEmbeddingsRAG