Ronin@DeRonin_
Practical tasks for the article "How to become AI engineer in 6 months"
it's been almost 4 days since I published the article, and realized one thing
you already have all the resources, but not everyone fully understands what exactly to do in practice
so I put together a list of practical tasks to help reinforce the knowledge from the article
Week 1: Python + Git + Terminal
1. CLI expense tracker (Python)
Write a program: add an expense, show the list, save it to a JSON file. Only built-in libraries (json, sys, argparse)
2. Weather script via API (HTTP, Python)
Get data from Open-Meteo (no API key needed). Parse the JSON, output temperature and precipitation in a readable format. Handle 404 and timeout
3. First GitHub repository (Git)
Push both projects above to GitHub with a proper README. Create a .gitignore, commit several versions with meaningful messages
4. CSV file analysis with SQL (Python)
Download any CSV (for example, sales data from Kaggle). Load it into SQLite via pandas, then answer 5 questions: top 5 categories, average, filtering by date
5. Async multi-request script (Async)
Write an async script that requests weather for 5 cities simultaneously using asyncio + httpx. Compare the execution time with a synchronous version
⏩---------------------------------------------------⏪
Week 2: FastAPI + first LLM
1. FastAPI service for the tracker (FastAPI)
Rewrite the CLI expense tracker into a REST API: POST /expense, GET /expenses, DELETE /expense/{id}. Use Pydantic models, uvicorn, test it via /docs
2. First LLM API call (LLM)
Connect to the OpenAI or Anthropic API. Write 5 different prompts for one task (for example, text summarization). Compare the quality of the outputs, which one is better and why
3. Invoice parser with structured output (LLM)
Give the LLM raw invoice text and get back a Pydantic object: invoice_number, amount, items[], due_date. Use Instructor or native structured output
4. Streaming responses via FastAPI (LLM, FastAPI)
Connect stream=True to the LLM and return the response through StreamingResponse in FastAPI. Check that tokens appear as they are generated, not all at once
⏩---------------------------------------------------⏪
these are at least the tasks you need to complete over the next 10-14 days
if this tweet gets enough feedback, I'll keep sharing more practical tasks
for each week or even each month, so you can go through this path from A to Z
stay focused pls...
