AI & LLM Engineering
A structured path from ML fundamentals to production AI systems. Built for software engineers who want to understand — and build with — modern AI, not just use it.
Machine Learning for Software Engineers
High-level understanding of supervised, unsupervised, and reinforcement learning. How models learn from data — training, validation, overfitting, and generalization.
Neural Networks — How They Actually Work
Layers, weights, activations, and backpropagation explained for engineers. Focus on the computation graph, not the calculus.
Tokenization & Context Windows
How text becomes tokens (BPE, WordPiece, SentencePiece), tokenizer vocabulary, context window limits and why they matter, positional encoding strategies.
Embeddings — Representing Everything as Vectors
How text, images, and structured data are converted into dense vector representations. Word2Vec, sentence embeddings, and why cosine similarity works.
Similarity Search Techniques
Distance metrics (cosine, euclidean, dot product), exact vs approximate nearest neighbor search, HNSW algorithm, FAISS library, and practical considerations for vector search at scale.
Data Pipelines for ML Systems
How data flows from raw sources to model training and inference. Feature engineering, data quality, and the practical reality of 'garbage in, garbage out.'