Learn Python.
Then learn to judge the code — even when AI writes it.
Rungo is a single climb from your first line of Python to reading, verifying and judging real code — including the stuff an AI generated. No track to pick. No wall of theory. Write code, run it, watch it pass, climb the next rung.
12 stages · 95 hands-on lessons · real Python in your browser, no setup.
# your first working program name = "world" print(f"hello, {name}")
Built for brains that bounce off normal courses
Most coding courses are a wall of theory and a punishing grind. ADHD brains stall on exactly that — the slow start, the no-feedback slog, the lost place. Rungo is engineered around how those brains actually learn. Good for them, better for everyone.
Run it, see it, feel it
Write code, hit run, watch it pass — instantly. That run-it-and-see-it-work hit is the dopamine loop, and coding is one of the few skills that delivers it on tap.
One bite-sized lesson
Every lesson is one concept and a ≈3-minute estimate on the label. Lowering the bar to START beats the biggest barrier there is — you only have to do one.
One-click resume
Come back after a day or a month and there's a single Continue button that drops you exactly where you left off. No re-finding your place, no friction.
Streaks and momentum
XP, skill bars, a daily streak and “you’ve done 3 in a row — one more?” nudges. Progress you can see, rewards that land, no hearts that punish a mistake.
Spaced recall
Quick 'remember this?' warm-ups resurface what you learned a few lessons back. Active recall, not re-reading — the thing that actually makes it stick.
Build real things
Boss-builds at the end of each stage — a Mad Libs machine, an AI-assistant core, a code-review harness. Projects you keep, not throwaway exercises.
The skill that survives
when AI writes the code.
Anyone can prompt an AI into a block of Python. Almost nobody can tell whether it's right. Rungo's last two stages teach the rarer, durable skill: read a diff, find the bug, spot the hallucinated API, catch the plausible-but-insecure pattern, and decide — accept, reject, or rewrite.
- Read code, not just write it — review a diff, find the missing edge case, judge the structure, give real feedback.
- Judge AI-generated code — spot the confident-but-wrong output and the hallucinated function before it ships.
- Prove it with a harness — write the test that demonstrates the bug, then make the accept/reject/revise call.
Nobody else is teaching this as the headline. It's the whole reason the ladder ends where it does.
def average(nums): return sum(nums) / len(nums) # ↑ crashes on an empty list — is that ok?
Do a lesson. Run it. Watch it pass. Climb.
No syllabus to read, no environment to set up. The whole loop is a real Python challenge you finish in minutes — then the rung lights up and the next one opens.
Get placed
A 30-second placement drops you on the right rung — a complete beginner starts at First Light; someone who already codes can test out of the early stages instead of grinding them.
Write & run real Python
Each lesson is one concept and a short brief. Write Python in the browser, run it instantly, see the tests go green. Stuck? The in-lesson mentor nudges, it doesn't hand you the answer.
Clear the gate, climb
Each stage ends on a boss-build you keep in your portfolio. Clear its gate and the next rung unlocks — all the way up to judging AI-written code.
One path. Zero to architect.
A single linear climb through the roles the industry actually hires for — from your first line of Python to judging AI-generated code. Indicative bands for context, the skills each stage builds, and no pretence it arrives overnight.
Your first working program is three minutes away.
Take the 30-second placement, land on the right rung, and write your first real Python line before your coffee's cold. Free to start — no card.