💪 PM Gym
Free practice platform for aspiring Product Managers.
Built for career switchers — analysts, engineers, designers, and consultants who want to break into PM without paying for a bootcamp.
🔗 Try it live →
What's Inside
| Level | Topics | What You'll Learn |
|---|
| 🌱 Beginner | 3 modules | What a PM does, product metrics, key frameworks |
| 📈 Intermediate | 3 modules | North Star metrics, A/B testing, prioritization (RICE, HEART, MoSCoW) |
| 🚀 Advanced | 3 modules | Product strategy, full case studies with model answers |
Every topic includes:
- 📖 A 3-min lesson with real company examples (Spotify, Airbnb, Slack, Duolingo)
- 🧠 5 quiz questions with explanations for every answer
- 🔗 3 curated free resources (no paywalls)
Advanced level adds:
- 📋 Real PM case studies — write your answer, then reveal a structured model answer and rubric
- Self-evaluation against dimensions like Problem Framing, Data Investigation, and Solution Quality
App features:
- 🔥 Streak counter — tracks consecutive days practiced
- 📊 Progress dashboard — completed topics and quiz scores
- 🌙 Dark / light mode
- 📱 Mobile-first, works on any device
- 💾 Progress saved locally — no account needed
Run Locally
No build step. No package installs. Just a browser.
git clone https://github.com/sanketjoshi2012/pm-gym.git
cd pm-gym
Then pick whichever option fits:
Option 1: Node (recommended if you have Node installed)
npx serve .
# Open http://localhost:3000
Option 2: Python (if you have Python 3 installed)
python3 -m http.server 4321
# Open http://localhost:4321
Option 3: No server needed
Just open index.html directly in your browser. Most things work fine without a server.
Tech Stack
Intentionally minimal:
- Vanilla HTML, CSS, JavaScript — no frameworks
localStorage for progress, streaks, and quiz scores- Google Fonts (DM Sans + Space Grotesk)
- Zero build tooling, zero backend, zero API calls
Contributing
PM Gym is community-driven. The most valuable contribution is new content — lessons, quiz questions, case studies, and resources.
You don't need to know how to code. If you can edit a text file and write clearly, you can add a topic.
→ Read CONTRIBUTING.md for templates and content standards.
Ideas for new topics:
- Intermediate: Stakeholder Management, Go-to-Market for PMs, User Research Methods
- Advanced: Meta Feed ranking drop, Notion onboarding redesign, DoorDash pricing strategy
Current experiment
Bet: does AI scoring on case answers make people practise again?
We will be adding an LLM grader that scores written case answers against the existing rubric, quotes the learner's own words as evidence, and says what a strong answer would add.
The question this answers: do people submit a second answer on the same case within 14 days?
Decision Rule
By end of week 4, among at least 25 people who submit one scored answer:
- Continue if 30% or more submit a second answer within 14 days, and the AI score agrees with my own grading on at least 16 of 20 test answers.
- Freeze if the repeat-submission rate is under 15%. Write up the case study, put it on the CV, move all evenings to job applications.
- One more iteration if between 15% and 30%: fix the single biggest drop-off visible in Mixpanel, then decide again. No third extension.
Versioning
See CHANGELOG.md for a full history of what's been added.
Current version: v1.3.6
License
MIT — free to use, share, and build on.
Built by Sanket Joshi. Contributions welcome.
