Computing & Artificial Intelligence
From foundational machine learning to modern generative architectures: self-attention, token sampling, and autonomous agent loops.
How AI Agents Work
The architectural loop behind autonomous planning, tool execution, and stateful problem solving
Inquiry Learning Paths
Structured sequences designed to take you from a familiar question to complete first-principles understanding.
How Computers Execute and Store Information
Progress from quantum band gaps in doped silicon to CMOS logic gates, binary adders, pipelined CPU instruction execution, coherent SRAM caches, and operating system kernels.
“How does purified quartz sand doped with boron and phosphorus become a microprocessor capable of executing billions of arithmetic instructions and running an operating system every second?”
How Semiconductors and Transistors Actually Work
Explains how doping silicon crystals creates p-n junctions and voltage-controlled MOSFET switches with zero moving parts.
How Binary and Logic Gates Became Computation
Shows how Claude Shannon mapped George Boole’s symbolic logic onto physical on/off switches (AND, OR, NOT, XOR) to represent numbers and decisions.
How CMOS Transistors Form Logic Gates
Details how complementary PMOS pull-up and NMOS pull-down transistor networks implement logic gates with near-zero static power dissipation.
How Binary Arithmetic Logic Units Actually Add Numbers
Combines XOR and AND gates into half-adders, full-adders, and carry-lookahead circuits that perform binary addition, subtraction, and comparisons in picoseconds.
How a CPU Executes an Instruction
Steps through the Fetch–Decode–Execute–Writeback pipeline, register files, branch prediction, and out-of-order execution inside a modern processor core.
How CPU Cache Hierarchies Overcome the Memory Wall
Explains why main DRAM is hundreds of times slower than the ALU—and how L1/L2/L3 SRAM caches and MESI coherence protocols prevent processor stalls.
How an Operating System Actually Runs Software
Reaches the software abstraction layer: how the OS kernel uses hardware timer interrupts, virtual-memory page tables, and system calls to multiplex applications safely.
How Compilers Translate Human Code into Machine Instructions
Completes the journey from human thought to silicon execution: how compilers lex, parse, optimize via SSA form, and allocate hardware registers to emit native machine code.
How the Internet Carries Information Around the World
Follow a network request across every physical and protocol layer: BGP autonomous systems, IP packet hops, submarine optical amplifiers, 802.11 Wi-Fi radio modulation, Anycast DNS, and TLS cryptography.
“When you tap a link or type a web address on your phone, how do your request packets travel through air as radio waves and across ocean floors as laser pulses to fetch a webpage in milliseconds?”
How the Internet Actually Works
Introduces the global network of networks: 115,000+ independent Autonomous Systems exchanging reachability paths via BGP-4 and forwarding 1,500-byte packets without central control.
How Packets Actually Travel Across the Internet
Follows an individual packet hop-by-hop: TCP segmentation, IPv4/IPv6 TTL decrementing, longest-prefix-match TCAM lookups, and dynamic rerouting around severed links.
How Fiber-Optic Cables Carry Light Across Oceans
Descends to the ocean floor where 99% of intercontinental traffic travels as 1550 nm infrared laser light through silica fibers amplified every 70 km by erbium-doped optical repeaters (EDFAs).
How Wi-Fi Actually Works
Explains the wireless last hop: how your router maps digital bits onto 2.4/5/6 GHz QAM radio subcarriers (OFDMA) and coordinates airtime via CSMA/CA collision avoidance.
How Cellular Networks Route Calls and Data
Extends wireless routing to mobile wide-area telecommunications: microsecond base station MAC scheduling, RRC connection handshakes, seamless inter-tower handovers, and GTP tunnels across LTE and 5G cores.
What Happens When You Type a Website Address
Connects packet transport to the application layer: hierarchical Anycast DNS resolution, the TCP three-way handshake, TLS 1.3 key exchange, and browser DOM rendering.
How Public-Key Cryptography Actually Works
Explains the mathematical miracle behind HTTPS and TLS: how two computers that have never met establish an encrypted channel over public fiber using Diffie-Hellman and elliptic-curve cryptography.
How AI Systems Reason and Act in the Real World
Trace modern artificial intelligence from transformer self-attention and next-token probability distributions to autonomous agent loops and structured tool execution.
“How does a neural network trained only to predict the next token become an autonomous software agent capable of invoking external APIs, querying databases, and executing multi-step plans?”
How Large Language Models Generate Text
Establishes the cognitive engine: subword BPE tokenization, high-dimensional vector embeddings, multi-head self-attention (Q, K, V matrices), and autoregressive next-token probability sampling.
How Transformers Changed AI
Explains the architectural revolution that solved sequential recurrence, detailing scaled dot-product attention mathematics, multi-head representation subspaces, and empirical neural scaling laws.
How AI Agents Work
Wraps the stateless language model inside an iterative runtime harness—introducing the Observe–Think–Act–Reflect control loop, working context windows, and persistent episodic memory.
How an AI Agent Decides Which Tool to Use
Shows how probabilistic token generation bridges to deterministic software execution via JSON schema constrained decoding, semantic tool routing, and host runtime validation.
All Published Explainers
Every verified first-principles analysis in this domain.
How AI Agents Work
The architectural loop behind autonomous planning, tool execution, and stateful problem solving
How Large Language Models Generate Text
From raw prompt strings to token embeddings, self-attention calculations, and next-token probability sampling
How Search Engines Actually Work
How search engines crawl the web, build inverted indexes, and rank billions of pages in 150 milliseconds using BM25, PageRank, and neural vector search
How an AI Agent Decides Which Tool to Use
The mechanics of schema matching, token probabilities, and semantic tool routing
Why AI Chatbots Sometimes Make Things Up
The mechanics of hallucination, probability vs. truth, and the engineering behind grounded AI responses
Why Search Results Differ Between People
Geographic IP Anycast routing, session history vectors, freshness ranking, and A/B experiment flight allocation
Why Neural Networks Were Invented
From McCulloch-Pitts threshold logic and the Rosenblatt perceptron to Minsky's XOR crisis and the backpropagation revolution
How Transformers Changed AI
From sequential recurrence and vanishing gradients to scaled dot-product attention, multi-head projections, and modern neural scaling laws
How Reinforcement Learning from Human Feedback (RLHF) Aligns AI
From Bradley-Terry preference modeling and PPO reward hacking to KL divergence penalties and direct preference optimization
How Mixture of Experts (MoE) Scales Large Language Models
From dense transformer bottlenecks and sparse gating to top-k routing, expert capacity, and Mixtral parameter scaling
How Vector Search and Dense Retrieval Work
From word embeddings and bi-encoder transformers to cosine similarity, HNSW small-world graphs, and product quantization
How AI Agent Memory Systems Work: From Working Context to Episodic Recall
From context window limits and virtual memory paging to vector recall, recursive summarization, and self-reflective memory consolidation
Technical Systems & Protocols Analyzed
Hardware, protocol switches, and central clearing houses examined in this hub.