Week 1 - AI, LLMs, Linux, Security

[AI Orientation] What is Artificial Intelligence?

AI is the field of building systems that perform tasks associated with human intelligence, such as reasoning or perception.

[AI Orientation] What is the difference between AI, ML and deep learning?

AI is the broad field; ML learns patterns from data; deep learning is ML using multilayer neural networks.

[AI Orientation] What is supervised learning?

Training on labeled examples to predict known target outputs, such as prices or categories.

[AI Orientation] What is unsupervised learning?

Finding patterns in unlabeled data, such as customer clusters.

[AI Orientation] What is reinforcement learning?

An agent learns actions through interactions and reward signals to maximize long-term reward.

[AI Orientation] What is generative AI?

AI that creates content such as text, images or code from learned patterns.

[AI Orientation] What is model hallucination?

A generated claim that sounds plausible but is unsupported or false.

[AI Orientation] What is bias in AI?

Systematic skew in data, models or outcomes that can create unfair or inaccurate results.

[LLM Platforms] What is a large language model?

A neural network trained on large text or code corpora to predict tokens and generate language-based outputs.

[LLM Platforms] What is a token?

A unit of text processed by a model, often a word fragment, word or punctuation mark.

[LLM Platforms] What is a context window?

The maximum token span a model can consider in one request, including prompts and generated output.

[LLM Platforms] What is inference?

Running a trained model to produce a prediction or generated response.

[LLM Platforms] What is temperature in LLM generation?

A sampling setting that influences how random or varied token choices are.

[LLM Platforms] What is an embedding?

A numeric vector representing semantic information for similarity search and retrieval.

[LLM Platforms] What is retrieval-augmented generation?

Retrieving relevant external information and adding it to a prompt to ground an LLM response.

[Prompt Engineering] What is prompt engineering?

Designing instructions and context to guide an AI model toward useful, reliable outputs.

[Prompt Engineering] What is zero-shot prompting?

Asking a model to perform a task without worked examples in the prompt.

[Prompt Engineering] What is few-shot prompting?

Providing a small number of examples to demonstrate the desired pattern or format.

[Prompt Engineering] What is prompt context?

Background information and constraints supplied to help the model answer appropriately.

[Prompt Engineering] Why specify output format in a prompt?

It reduces ambiguity and helps produce consistent, usable responses.

[Prompt Engineering] What is prompt injection?

Untrusted content attempting to override a model's authorized instructions.

[Prompt Engineering] How do you evaluate an LLM answer?

Check factual accuracy, relevance, completeness, safety and adherence to the requested format.

[Linux and Scripting] What is Linux?

A family of operating systems built around the Linux kernel, widely used for servers and development.

[Linux and Scripting] What does pwd do?

Prints the current working directory.

[Linux and Scripting] What do ls and cd do?

ls lists directory contents; cd changes the working directory.

[Linux and Scripting] What do grep and find do?

grep searches text for patterns; find locates files and directories matching conditions.

[Linux and Scripting] What does chmod do?

Changes file permissions, such as read, write and execute bits.

[Linux and Scripting] What is a shell script?

A file containing shell commands that automate repetitive tasks.

[Linux and Scripting] What is a pipe in Linux?

The | operator sends one command's standard output to another command's standard input.

[Linux and Scripting] What is an environment variable?

A named value passed to processes for configuration, such as PATH.

[Linux and Scripting] What does sudo do?

Executes a permitted command with elevated privileges, typically as root.

[Security Fundamentals] What is the CIA triad?

Confidentiality, Integrity and Availability are three core information-security goals.

[Security Fundamentals] What is authentication versus authorization?

Authentication verifies identity; authorization determines what an identity may access or do.

[Security Fundamentals] What is least privilege?

Giving identities only the permissions necessary for their tasks.

[Security Fundamentals] What is encryption?

Transforming plaintext into ciphertext using a key so unauthorized parties cannot read it.

[Security Fundamentals] What is hashing?

A one-way mapping from input to a fixed-size digest, used for integrity checks and password storage with suitable algorithms.

[Security Fundamentals] What is phishing?

Social engineering that tricks people into disclosing secrets or taking harmful actions.

[Security Fundamentals] What is multi-factor authentication?

Requiring independent factors, such as a password and an authenticator, to verify identity.

[Security Fundamentals] What is SQL injection?

An attack that manipulates SQL by mixing untrusted input into query text; parameterized queries help prevent it.