Learn something new every day
Here's one thing to learn today based on a topic you pick
What you will receive
Today's learning: Python
just now
Did you know?
Python's "walrus operator" (:=) lets you assign and use a value in one expression.
Example:
if (n := len(data)) > 10:
print(f"Too much data: {n} items")
Try it: Refactor one of your if statements to use :=
Source: PEP 572How it works
- 1You pick a topic you want to learn about
- 2AI finds an interesting, bite-sized concept
- 3You get one actionable learning each day
You configure
Python, JavaScript, machine learning, etc.
What do you want to learn about?
intermediate
beginner, intermediate, or advanced
sk-...
For generating learning content
View Python code
import requests
import os
from state import state
import random
TOPIC = os.environ.get("LEARNING_TOPIC", "programming")
LEVEL = os.environ.get("SKILL_LEVEL", "intermediate")
OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY")
# Track what we've taught to avoid repetition
taught = state.get("taught_concepts", [])
prompt = f"""Teach me ONE interesting {LEVEL}-level concept about {TOPIC}.
Previously taught (avoid these): {', '.join(taught[-10:]) if taught else 'None yet'}
Format:
- Concept name: [name]
- Explanation: [2-3 sentences]
- Example: [code or practical example]
- Try it: [one small exercise]
Keep it concise and actionable. Under 100 words total."""
response = requests.post(
"https://api.openai.com/v1/chat/completions",
headers={"Authorization": f"Bearer {OPENAI_API_KEY}"},
json={
"model": "gpt-4o-mini",
"messages": [{"role": "user", "content": prompt}],
"max_tokens": 200
}
)
lesson = response.json()["choices"][0]["message"]["content"]
# Extract concept name and track it
lines = lesson.split("\n")
concept = lines[0].replace("Concept name:", "").replace("-", "").strip() if lines else TOPIC
taught.append(concept)
state["taught_concepts"] = taught[-20:] # Keep last 20
print(lesson)Humrun remembers what concepts it taught you to avoid repetition.
Suggested schedule: Every weekday at 9 AM•Notifications: After every run