Module 1

Python for
AI Automation
Complete Roadmap · Day-by-Day Schedule · Resources · Projects

Tailored for: Comfortable Python level · 1–2 hours/day · Real Estate | Finance | E-Commerce

3 ~35 7 3
Weeks Total Total Hours New Skills Final Projects
 ■ Module Overview
What you'll learn and why it matters

What You Already Know (Skip These)

Since you're comfortable with OOP and pandas/numpy, we skip Python basics entirely. This module focuses ONLY on
the gaps between your current level and what's needed to build real AI automation systems.

OOP & Classes ■ Already know — skip

pandas / numpy ■ Already know — skip

loops, functions ■ Already know — skip

File I/O basics ■ Already know — skip

What You'll Master in This Module

requests / httpx Calling any web API from Python — the backbone of automation

JSON handling Parsing and building API responses cleanly and reliably

python-dotenv Securing API keys — never expose secrets in code

asyncio Running multiple agents simultaneously for 10x speed

FastAPI Wrapping your automation as a deployable web service

Error handling Retries, timeouts, fallbacks — production-grade resilience

Scheduling Running automations on a timer without manual triggering

Why These Skills Matter

Every AI automation — whether it's a real estate chatbot, a finance report agent, or an e-commerce monitor — is
ultimately Python talking to APIs. Master these 7 skills and you can build ANY automation that exists. They are the
universal building blocks.

Module 1: Python for AI Automation · Page 2

 Week 1 — APIs & Data Flow
Days 1–7 · ~10–12 hours total

This week you go from 'I know Python' to 'I can talk to any service on the internet.' APIs are the nervous system of every
AI automation. Master them first.

DAY-BY-DAY SCHEDULE

D1 REST APIs & HTTP

Learn GET/POST/PUT/DELETE · status codes · headers · Postman

D2 requests Library
Install requests · GET calls · handle JSON response · print data

D3 API Authentication
API keys · Bearer tokens · OAuth basics · store keys safely

D4 POST Requests & Payloads

Send data to APIs · JSON body · form data · real API practice

D5 python-dotenv
Create .env file · load secrets · .gitignore it · best practices

D6 Error Handling
try/except · status code checks · retries with backoff · timeouts

D7 Practice Day
Build: Property data fetcher using a free real estate / geo API

KEY CONCEPTS

requests Library — Core Patterns

The requests library lets you call any API in 2–3 lines. You'll use this in literally every AI project you build:

[Link](url) Fetch data — reading property listings, stock prices

[Link](url, json={}) Send data — submitting to AI APIs, creating records

[Link]() Parse JSON response into a Python dictionary

response.status_code Check if call succeeded (200) or failed (4xx/5xx)

headers={'Authorization':…} Pass API keys securely in every request

python-dotenv — Never Expose API Keys

One of the most important professional habits. Your .env file holds all secrets and is NEVER committed to GitHub:

.env file OPENAI_API_KEY=sk-abc123 (never share this file)

.gitignore Add .env here so Git never tracks it

load_dotenv() Loads .env values into environment at runtime

[Link]('KEY') Retrieves the secret value safely in your code

Week 1 Mini-Challenge

Module 1: Python for AI Automation · Page 3

By end of Day 7 you should be able to: write a Python script that calls a free API (e.g. OpenWeatherMap, REST Countries, or
any real estate data API), extracts specific fields from the JSON response, and prints a formatted summary. If you can do this
comfortably — Week 1 is complete. ■

Module 1: Python for AI Automation · Page 4

 Week 2 — Async, FastAPI & JSON Mastery
Days 8–14 · ~10–12 hours total

This week separates hobbyists from professionals. Async programming makes your agents run 5–10x faster. FastAPI
lets you ship your automation as a real service clients can use.

DAY-BY-DAY SCHEDULE

D8 JSON Deep Dive

Nested JSON · loops over lists · .get() safely · write JSON to file

D9 httpx & Async Basics

Install httpx · async/await syntax · event loop concept

D10 Async API Calls

async def · await [Link]() · run multiple calls at once

D11 [Link]()
Call 5 APIs simultaneously · measure speed vs sequential

D12 FastAPI Basics

Install FastAPI · create first endpoint · run with uvicorn

D13 FastAPI + Pydantic

Request/response models · input validation · auto docs at /docs

D14 Practice Day

Build: Async price checker hitting 5 e-commerce APIs at once

KEY CONCEPTS

asyncio — Why It's a Game Changer

Without async: calling 10 APIs takes 10 seconds. With async: it takes 1 second. Your AI agents WILL need to do many
things at once:

async def function() Marks a function as asynchronous — can be paused/resumed

await something() Pauses here and lets other tasks run while waiting

[Link](*tasks) Run ALL tasks simultaneously — massive speed boost

[Link](main()) Entry point — starts the async event loop

[Link]() Async-compatible HTTP client — use instead of requests

FastAPI — Your Agent's Front Door

FastAPI turns your Python automation into a service that anything can call — n8n, websites, mobile apps, other agents:

@[Link]('/listings') Create an endpoint — your agent responds to HTTP calls

@[Link]('/analyze') Receive data, process it with AI, return result

BaseModel (Pydantic) Define exact shape of input/output — auto-validated

/docs endpoint FastAPI auto-generates interactive API documentation

uvicorn main:app Run your API server — ready to deploy in 1 command

Module 1: Python for AI Automation · Page 5

Week 2 Mini-Challenge
Build a FastAPI app with one POST endpoint that receives a city name, calls a weather API and a property price API
simultaneously using [Link](), and returns a combined JSON response. If it works — Week 2 is complete. ■

Module 1: Python for AI Automation · Page 6

 Week 3 — File Handling, Scheduling & OpenAI API
Days 15–21 · ~10–12 hours total

The final week bridges everything into actual AI. You'll handle real business files (PDFs, Excel, CSVs), automate
recurring tasks, and make your first OpenAI API call — the moment your Python becomes genuinely intelligent.

DAY-BY-DAY SCHEDULE

D15 File Handling for AI

Read/write CSV, Excel (openpyxl), JSON files · pathlib

D16 PDF Processing

Extract text from PDFs with pdfplumber · parse invoices

D17 schedule Library

Run functions every hour/day/week · background tasks

D18 OpenAI API Setup

Get API key · pip install openai · first completion call

D19 OpenAI Chat Completions

Messages array · system/user roles · parse AI response

D20 Connecting It All

API call → AI analysis → save result → schedule to repeat

D21 Final Project Day

Build one of the 3 capstone projects below — full build

KEY CONCEPTS

File Processing Libraries

openpyxl Read and write Excel files — perfect for finance/real estate reports

pdfplumber Extract text and tables from PDFs — invoices, statements, contracts

[Link] Modern file path handling — works on Windows, Mac, Linux

csv module Built-in CSV reading/writing — property data, transaction records

json module Built-in JSON handling — API data, config files, agent memory

schedule Library — Automation on Autopilot

[Link]().day Run a task once every day at a set time

[Link]().hour Run every hour — live price monitoring

[Link]().monday Run every Monday — weekly report generation

schedule.run_pending() Place in while loop to keep scheduler alive

OpenAI API — First Contact with AI

[Link] The main call — send messages, receive AI response

system message Give AI its role: 'You are a real estate analyst'

Module 1: Python for AI Automation · Page 7

user message The actual question or data you want analyzed

[Link][0] Extract the AI's text reply from the response object

model='gpt-4o-mini' Start with mini — cheap, fast, surprisingly capable

Module 1: Python for AI Automation · Page 8

■ Courses & Resources
Curated — No Fluff. Only What You Need.

FREE COURSES & TUTORIALS

Harvard's free Python course. Even at your level,

■ CS50P — Python [Link]/python FREE
weeks 4–9 are gold for file I/O and APIs.

■■ Corey Schafer Best Python tutorials on YouTube. Watch his

[Link]/@coreyms FREE
YouTube requests, OOP, and decorators playlists.

Article-based deep dives. Search: 'requests

■ Real Python [Link] FREE
tutorial', 'asyncio', 'FastAPI'. All free.

■ FastAPI Official The official tutorial is genuinely excellent. Follow it

[Link]/tutorial FREE
Docs top to bottom — takes ~3 hours.

■■ TechWithTim Practical Python projects — API tutorials,

[Link]/@TechWithTim FREE
YouTube automation scripts, FastAPI builds.

Official examples for every API feature. Bookmark

■ OpenAI Cookbook [Link] FREE
this — you'll use it for months.

■■ Akamai Async The definitive guide to asyncio in Python. Read [Link]/async-io-pytho

FREE
Tutorial this once — understand async forever. n

■ Automate Boring Free book online. Chapters 12-18 cover files,

[Link] FREE
Stuff Excel, PDFs, scheduling — exactly Module 1.

PAID COURSES (WORTH IT)

■ Udemy: 100 Days Best overall Python course. Even at your level, the
[Link] — Angela Yu PAID
of Code API and web sections are excellent.

■ [Link]: Andrew Ng's free short course on using OpenAI

[Link] FREE
ChatGPT API API. Only 1 hour. Do this in Week 3.

■ Arjan Codes Advanced Python patterns — clean code, async,

[Link]/@ArjanCodes FREE
YouTube FastAPI architecture. Elevates your coding.

DOCUMENTATION TO BOOKMARK

requests docs [Link] — complete reference, examples for everything

httpx docs [Link] — async HTTP, all features documented clearly

FastAPI docs [Link] — best API docs you'll ever read, with examples

openai-python [Link]/openai/openai-python — official Python SDK + examples

pydantic docs [Link] — data validation, used everywhere in AI systems

schedule docs [Link] — simple, clear scheduling reference

Module 1: Python for AI Automation · Page 9

 ■ Final Projects
3 Projects — One Per Niche — Build on Day 21

These projects use EVERY skill from the 3 weeks. Each is designed to be genuinely useful — something you could show
to a real client today.

Real Estate Property Alert System ■■ Medium 3–4 hrs

01 ■ Real Estate

What It Does
A scheduled Python agent that monitors property listings via API, filters by criteria (price, location, bedrooms), uses
OpenAI to write a natural-language summary of each new property, and sends a formatted email digest every morning.

Skills Used

requests Fetch listings from Zillow / RapidAPI real estate endpoint

python-dotenv Store API keys for real estate API + OpenAI + email

OpenAI API Generate natural language property summaries automatically

schedule Run every morning at 8 AM without manual triggering

openpyxl Save all listings to Excel for client reference

smtplib Send formatted email digest to agent's inbox

Why It's Valuable

Real estate agents spend 2+ hours daily checking listing sites. This does it in 30 seconds. A local agent would pay
$200–$500/month for this. Build it → show them → get your first client.

Finance Expense Categorizer & Report ■■ Medium 3–4 hrs

02 ■ Finance

What It Does
A FastAPI service that accepts a CSV or PDF bank statement upload, uses OpenAI to automatically categorize every
transaction (food, rent, utilities, etc.), calculates monthly spending summaries, and returns a clean Excel report with
charts.

Skills Used

FastAPI POST /analyze endpoint accepts file upload from any client

pdfplumber Extract transaction rows from PDF bank statements

pandas Group and aggregate transactions by category

OpenAI API Categorize each transaction description intelligently

openpyxl Generate Excel report with formatted tables

pydantic Validate incoming request structure automatically

Why It's Valuable

Module 1: Python for AI Automation · Page 10

 Accountants manually categorize hundreds of transactions per client. This does it in seconds. Show this to a local accounting
firm — they'll want it immediately. Charge $300 setup + $99/month per client.

E-Commerce Competitor Price Monitor ■■■ Hard 4–5 hrs

03 ■ E-Commerce

What It Does
An async Python agent that monitors competitor product prices across multiple e-commerce sites simultaneously, detects
price drops or increases, uses OpenAI to write recommended pricing actions, and sends a real-time WhatsApp/email
alert when a competitor changes price on a tracked product.

Skills Used

httpx + asyncio Monitor 10 competitor URLs simultaneously — fast

schedule Run checks every 30 minutes around the clock

OpenAI API Generate pricing recommendations based on market data

python-dotenv Store API keys for OpenAI + Twilio WhatsApp

requests (Twilio) Send WhatsApp alert when price change detected

openpyxl Log all price history to Excel for trend analysis

Why It's Valuable

Every e-commerce seller needs to know what competitors charge. This is async so it checks 10 sites in the same time as
checking 1. E-commerce businesses would pay $500–$2,000/month for a live competitor monitoring tool.

Module 1: Python for AI Automation · Page 11

 Module 1 Completion Checklist

■ ■ Can call any REST API with requests and handle JSON response
■ ■ API keys are always stored in .env — never hardcoded
■ ■ Can run multiple async API calls simultaneously with [Link]()
■ ■ Have a working FastAPI endpoint deployed locally
■ ■ Can extract data from PDFs and write to Excel with Python
■ ■ Have a scheduled automation running on a timer
■ ■ Have made at least one OpenAI API call that returns useful output
■ ■ One of the 3 final projects is fully working and on GitHub

What's Next → Module 2: Prompt Engineering

Once you complete all 3 projects, you're ready for the next module

You now speak the language of the internet. Every API, every service, every AI model is accessible
to you. Module 2 teaches you to speak the language of AI itself.

Build fast. Ship ugly. Learn everything. ■

Module 1: Python for AI Automation · Page 12