Embeddings and Similarity — How Machines Understand Text Similarity
From Word2Vec to sentence-transformers — what embeddings are, how cosine similarity works, and why semantic similarity is not the same as keyword matching.
Read more →From Word2Vec to sentence-transformers — what embeddings are, how cosine similarity works, and why semantic similarity is not the same as keyword matching.
Read more →A detailed computational example of backpropagation with manual calculations. From forward pass through backward pass to the optimizer step.
Read more →Detailed computational example of the attention mechanism from 'Attention is All You Need'. Step by step through embeddings, Q, K, V matrices to final attention weights.
Read more →A practical guide to tokenizers in large language models. Comparison of BERT, GPT-4, GPT-2, T5, StarCoder and XLM-RoBERTa with multilingual examples, plus building your own tokenizer from scratch.
Read more →Comprehensive explanation of Java ORM ecosystem - what are JPA, Hibernate, EntityManager, Session, JPQL, HQL and Criteria API and how they relate to each other.
Read more →Analysis of the N+1 problem in ORMs, its causes and effective solutions with practical examples in Hibernate.
Read more →Second post in the Kubernetes series – adding Helm charts to the simple cluster from the previous article.
Read more →First post in the Kubernetes series – building a simple cluster from scratch.
Read more →