I’m Andrey Gein

• London-based → in Bremen for only a few Saturdays during the semester
• I love Python!
• Telegram: @andgein
• I love Python!!! Ask me any questions about Python!!111
• I used to be a software engineer, and now I’m an SRE
• During my life and career, I’ve been using not only Python, but also C++ and C#, Java and JavaScript, Go, PHP, Perl and Bash, Delphi, Kotlin, Assembly, SQL,
  Ruby, Erlang, Objective-C, even Haskell, Prolog, R and Basic…

It’s me!

What is this course about?

• It’s about Python! And Python is the best my favourite programming language!
• I’m not going to teach you programming in general
• Let’s focus on Python
  • Its model and idioms
  • The standard library and ecosystem
  • Software architecture, code quality, testing, debugging
• I’m not going to teach basic Python syntax — please use documentation, LLMs, Google, …
• Despite that, you must know all Python syntax by the end

You can read Python!

• I can show you a Python function

  def fahrenheit_to_celsius(fahrenheit: float) -> float:
      return (fahrenheit - 32) * 5 / 9

  and you understand it, don’t you?

• avg = sum(numbers) / len(numbers)

  You get the idea, right?
• One more function:

  def cointoss(n,t):
      return (lambda a:"\n".join(str(i)+":\t"+"*"*a.count(i) for i in range(min(a),max(a)+1)))\
          ([sum(randint(0,1) for _ in range(n)) for __ in range(t)])

Our goals

After the course, I’d like you to be able to

• open any Python code in the world and understand what it does
• find a bug, write tests, suggest improvements
• suggest code or architecture improvements
• distinguish Pythonic code from non-Pythonic code
• know when Python is a good choice and when it’s not
• understand how Python (and CPython) works at a low level
• know details of Python tooling: PyPI, venv, uv, …
• feel comfortable talking about Python libraries: requests, HTTPX, Django, FastAPI, pandas, NumPy & SciPy, YAML, Pydantic, Black & Ruff & Pylint, …

Why learn Python when we have LLMs?

• Likely the most controversial topic in education today
• I can’t predict the future, but I’m seeing how LLMs and agents have changed the industry so far
• And we still need a smart human in the loop
• Geoffrey Litt: Understanding is the new bottleneck [text] [video]
• Make the most of AI tooling
• … but build your own understanding of concepts and your ability to read code and support systems

Course structure

• Lectures: September 5th, October 17th, November 14th
• Seminars: once a week, Friday at 11:15
• Homework: 3 practical assignments per semester
• Labs: large tasks taking 3–4 weeks with a face-to-face walkthrough at the end
• Exam: at the end
• Everything above is mandatory to pass the class. Everything below is optional
• Conference: choose a topic, plan and prepare the material, give a talk (late November)
• Team contest: a Saturday competition in late October with no algorithmic or data-structure tasks, but it’ll be fun — details later, teams of three

Grading

• Exam: 100 points
• Bonus points: 20 for giving a talk at the conference, 10 for quizzes during seminars
• Practice: 100 points during the semester
• Python Gym: 10
• Homeworks: 3 × 10
• Labs: 2 × 30
• Bonus points: 20 for participation and winning the team contest

LLM usage

• General rule: please use all the modern technologies available to you
• In particular, you can freely use LLMs and agents in Labs, when preparing a Talk, and in the Contest
• You still need to understand everything in your Lab’s code and be able to present it
• For Python Gym, seminars, and homework, you can ask an LLM any questions or search the internet
• Ask “Why are strings immutable in Python?” rather than “Write code for this task”

Pythonic

if items:
    print("There are items")

if len(items) > 0:
    print("There are items")

for name in names:
    print(name)

for i in range(len(names)):
    print(names[i])

squares = []

for x in numbers:
    squares.append(x * x)

squares = [x * x for x in numbers]

Zen of Python, by Tim Peters

import this