Python basics, lists, dictionaries, and sets

🔴 MUST KNOW · Learn enough to operate AI services; keep existing MERN responsibilities in TypeScript when appropriate.

Overview

• Python is a dynamically typed language widely used by AI libraries.
• A list is an ordered collection, a dictionary maps keys to values, and a set stores unique hashable values.
• Python uses indentation to define code blocks rather than braces.

Why this matters

Many document-processing, model-training, numerical, and evaluation libraries are easiest to use in Python.

An everyday example

A list is a queue of documents; a dictionary is a labeled filing cabinet; a set is a guest list without duplicates.

A technical example

chunks = ["policy", "claim"]; metadata = {"tenant_id": "t1"}; unique_ids = {"C1", "C2"}.

How it works

1. Create values
2. Use indexing or key lookup
3. Loop or use comprehensions
4. Pass results to functions.

• Lists and dictionaries are mutable; sharing them can create accidental changes.

Architecture and flow

Parsed documents → lists of chunks → dictionaries of metadata → unique IDs for deduplication.

The insurance assistant example

A worker extracts policy paragraphs, attaches source metadata, and deduplicates document identifiers.

When to use it

For Python-facing data preparation and AI libraries.

When to avoid it

Do not rewrite a working Node service merely to use a hosted LLM API.

Interview answer

“I use Python where its AI ecosystem provides practical benefits and keep collection types explicit so ingestion and evaluation code remain readable.”

Benefits, trade-offs, and production considerations

• TypeScript interfaces and Python type hints do not provide runtime validation.
• Python's None, True, False, and, or, and not correspond roughly to JS's null, booleans, and logical operators, but truthiness and equality rules differ.

# No third-party dependencies: basic data preparation.
documents = [
    {"id": "p1", "text": "Claim reporting instructions", "active": True},
    {"id": "p2", "text": "Old wording", "active": False},
]
active_texts: list[str] = [d["text"] for d in documents if d["active"]]
seen_ids: set[str] = {d["id"] for d in documents}
for index, text in enumerate(active_texts):
    print(index, text)