Skip to content

Concepts - Bite-Size Topic Pages

Single-topic explanations. 5-10 minute reads. No exam scaffolding.

Read one when you keep seeing a term and want to actually understand it.

How these are organized

Each page covers one concept: - What it is, in plain English - Why it exists (the problem it solves) - A small concrete example - What to look at next

These are not exam-objective summaries. They're "explain it like I'm a smart engineer who just hasn't been exposed to this yet."

If you want exam coverage, head back to the Study Hub.

Brand new to all of this? Start with Day One - terminal, git, HTTP, and servers - then come back here once that vocabulary feels familiar.

Want a curated path through these? Try Cloud from Scratch or AI from Scratch, which order the concepts into 8-phase paths.

Want them grouped by subject across the whole repo? See the topic indexes.

Cloud Computing

Topic Read when...
What is cloud computing? You want the foundational mental model
IaaS vs PaaS vs SaaS The three terms keep coming up
Regions and availability zones You're asking "where does my code actually run?"
Shared responsibility model You're wondering "what's AWS's job vs mine?"
VPC explained Networking inside the cloud confuses you
IAM explained "Roles vs policies vs federation" is fuzzy
Serverless explained You keep hearing "serverless" without a clear definition
CDN explained You see CloudFront/CloudFlare/Akamai and wonder what they do
Caching explained You want to know why caches are fast and why they lie
Load balancing explained "Layer 4 vs layer 7" keeps coming up and you want specifics
Autoscaling explained You want capacity to follow demand without an outage
Cloud cost basics The bill went up and nobody decided to spend more

Containers and Orchestration

Topic Read when...
Containers vs VMs You don't know why containers replaced VMs
Kubernetes in 10 minutes Everyone uses K8s; you want to know why

DevOps / Infrastructure

Topic Read when...
Terraform explained You want to understand IaC without diving into HCL
CI/CD explained "Pipelines" is hand-waved everywhere; you want specifics
Observability basics "Logs vs metrics vs traces" needs a one-page answer
Idempotency explained You're worried about retries doing things twice
Eventual consistency A read returned stale data and you want to know why
Queues vs streams You can't pick between SQS and Kinesis
Deployment strategies Rolling, blue-green, canary: which and when

Data and Databases

Topic Read when...
SQL vs NoSQL You're choosing a database and the options blur together
Eventual consistency A read returned stale data and you want to know why
Caching explained You want to know why caches are fast and why they lie

Networking and Security

Topic Read when...
DNS explained You want to understand how the internet finds things
TLS and HTTPS You want to know what the lock icon actually means
Secrets management You need credentials in an app without leaving copies everywhere

AI / Machine Learning - foundations

Topic Read when...
LLM basics You want to understand what a language model actually does
Transformer architecture You're ready to look inside the model
Embeddings and vector search You want to understand semantic search
Context windows and management You're hitting token limits in production
Multimodal models You want to use images / audio / video in prompts

AI / Machine Learning - building with LLMs

Topic Read when...
Prompt engineering You want to write prompts that actually work
Tool use and function calling You want the model to call your code
MCP explained "Model Context Protocol" keeps coming up
Structured outputs You want guaranteed-valid JSON from the model
RAG explained "Retrieval-augmented generation" is everywhere
Fine-tuning vs RAG You want to know which one to reach for
AI agents explained You're trying to figure out what "agent" really means
Agentic loops You want the deep-dive on the loop, not just the concept
Prompt caching Your input costs are higher than they should be

AI / Machine Learning - operations

Topic Read when...
Evals for LLMs You're shipping LLM features and wondering how to test them
Guardrails and safety You're worried about prompt injection or unsafe output
Prompt injection explained You want to know why prompt injection has no clean fix
AI threat modeling You're designing an AI feature and want to find the risk early
Inference servers You're considering self-hosting an open-weights model
Quantization and distillation You want a smaller, faster model
GPUs for AI You want to know what VRAM, FLOPS, and bandwidth actually decide

Companion docs

Suggest a topic

Missing a concept you keep getting asked about? Open an issue or PR. Criteria for new pages: - One topic, focused - 5-10 minute read - Plain English, no exam framing - Concrete example included

Contributing Guide