The book is the masterclass. This repository is the practice lab.
The book teaches the full reasoning, chapter walkthroughs, trade-off narratives, and interview storytelling. This repository helps you practice that thinking through templates, prompts, rubrics, diagrams, and companion artifacts.
This glossary defines common acronyms and terms used across the book and repository. It is a quick reference. The book explains these terms in system context.
API — Application Programming Interface: A contract that allows software systems to communicate.
SLA — Service Level Agreement: A formal commitment, often contractual, about service availability or performance.
SLO — Service Level Objective: A target reliability or performance goal used internally to manage a service.
RTO — Recovery Time Objective: How quickly a system must recover after a failure.
RPO — Recovery Point Objective: How much data loss is acceptable after a failure.
CQRS — Command Query Responsibility Segregation: A pattern that separates write operations from read operations.
DR — Disaster Recovery: The strategy for recovering systems after major outages or disasters.
HA — High Availability: Designing systems to remain available despite failures.
Idempotency: The property that repeating the same operation produces the same intended result, preventing duplicate side effects.
Backpressure: A mechanism that prevents overwhelmed services from accepting more work than they can handle.
Circuit Breaker: A resilience pattern that stops calls to a failing dependency to prevent cascading failure.
Saga: A pattern for managing distributed transactions through a sequence of local transactions and compensating actions.
Outbox Pattern: A pattern that stores events in the same transaction as business data so messages can be reliably published later.
CDN — Content Delivery Network: A distributed network of edge servers that serves content closer to users to reduce latency and improve scalability.
ABR — Adaptive Bitrate Streaming: A technique where the player switches between different video quality levels based on bandwidth, device capability, and buffer health.
HLS — HTTP Live Streaming: A streaming protocol that delivers video in small HTTP-based segments, widely used across devices.
DASH — Dynamic Adaptive Streaming over HTTP: An adaptive streaming standard that lets players select video quality from available segments.
DRM — Digital Rights Management: Technology that protects licensed content by controlling playback, decryption, downloading, and sharing.
Manifest: A playlist-like file that tells a video player which segments, bitrates, captions, and tracks are available.
Transcoding: Converting media into different formats, resolutions, or bitrates for different devices and networks.
Egress: Data transferred out of a cloud, storage, or CDN environment, often a major cost driver.
AI — Artificial Intelligence: Software techniques that enable systems to perform tasks associated with reasoning, prediction, language, perception, or decision support.
ML — Machine Learning: A branch of AI where systems learn patterns from data.
LLM — Large Language Model: A model trained on large text corpora to generate, summarize, classify, or reason over language.
RAG — Retrieval-Augmented Generation: A pattern where a language model uses retrieved documents or data to ground its response.
Embedding: A numerical representation of text, image, or data that captures semantic meaning for search or comparison.
Vector Database: A database optimized for storing and searching embeddings.
Graph Database: A database that represents entities and relationships as nodes and edges.
GraphRAG: A retrieval approach that combines graph relationships with retrieval-augmented generation to improve context and explainability.
Prompt Injection: An attack where a user or document attempts to manipulate an AI system’s instructions or behavior.
Reranking: Reordering retrieved results to improve relevance before presenting them to a model or user.
Token Budget: The limit on how much text a model can process or generate, often affecting cost and latency.
Kafka: A distributed event streaming platform often used for high-throughput messaging and event pipelines.
Cache Stampede: A failure mode where many requests hit the backend at once after cached data expires.
Hot Key: A key or partition receiving disproportionate traffic, causing imbalance.
Sharding: Splitting data across multiple partitions or nodes to improve scale.
Replication: Copying data across nodes or regions for availability, durability, or performance.
Eventual Consistency: A consistency model where replicas may temporarily differ but converge over time.
More terms can be added as the repository evolves.