AI Core Math Review
The core math neural networks rely on: scalars, vectors, dot products, and matrices, the toolbox behind attention and embeddings.
Start reading →Tutorials
A standing note on anything time-sensitive. Some of these series name specific cloud provider services, VC firms, accelerator terms, or product capabilities. All of that changes fast, so treat any such name, number, or feature list as a snapshot to verify, not a permanent citation. Series that touch funding, equity, or compliance (SAFEs, cap tables, SOC 2/HIPAA/GDPR, and similar) are a practical explanation of how these things generally work, not legal or financial advice.
Colab Math Practice
A hands-on companion to AI Core Math Review with the same scalar, vector, dot product, and matrix examples, runnable and editable in a Colab notebook.
Open in Colab →Understanding Transformers
A 16-part, intuition-first tour of how GPT-style models work, from learning by prediction to next-token generation, then to the hardware they run on. Pair it with the companion AI Core Math Review series if the underlying math is rusty, and Build Your Own Transformer for a hands-on notebook.
Start reading →Build Your Own Transformer
A hands-on companion to Understanding Transformers — assemble a working transformer yourself in a guided Colab notebook, in the order you would build one.
Open in Colab →Agentic AI
What turns a language model into an agent: a reasoning loop, tools it can call, and a goal it works toward across several steps instead of one reply, plus the frameworks (LangGraph, AutoGen, CrewAI, and others) built to orchestrate them and the safeguards that go around the loop in production.
Start reading →Colab Agent Practice
A hands-on companion to Agentic AI: build the loop, tool calling, planning, memory, and a small multi-agent team around a small open model, then rebuild the team in LangGraph, in an editable Colab notebook.
Open in Colab →Startups: The Ecosystem and How They're Funded
What a startup is, the stages it passes through and the risks that dominate at each one, where its capital comes from, how it's priced and structured, what founders give up to get it, the growth investors expect, where startups are built, and how cloud and AI changed the economics. The business grounding behind Core Cloud Architecture and the funded companies examined in Cloud and AI Architecture: Case Studies.
Start reading →Core Cloud Architecture
The building blocks of cloud systems (regions and networking, compute and storage, databases, caching, messaging, reliability, identity, security, observability, and delivery), with AWS, GCP, and Azure comparisons throughout. The technical grounding behind Cloud and AI Architecture: Case Studies.
Start reading →Generative AI Architecture
How generative AI systems are built and run: foundation models, prompting, embeddings, retrieval-augmented generation, vector databases, fine-tuning, evaluation, and the GPU/TPU infrastructure underneath. Pairs with Understanding Transformers for the model mechanics and Agentic AI for agents.
Start reading →Cloud and AI Architecture: Case Studies
Three fictional startups — Casewell (pre-seed/seed), Sentrio (Series A), and Virelane AI (late-stage/growth) — each get a full AWS, GCP, and Azure architecture in one article, a capability comparison table, architecture decision records, and due diligence questions. Applies Startups: The Ecosystem and How They're Funded, Core Cloud Architecture, and Generative AI Architecture together, end to end.
Start reading →