The local-first terminal workspace for technical interview prep.
Search 3,310 questions, browse 17,931 mappings across 662 companies, and track progress—all in the CLI.

---

⚡ Quickstart

pip install dooma
dooma

Using Dooma on Windows? See the Windows terminal troubleshooting guide for help with missing icons, text encoding, and narrow layouts.

See Dooma in Action

---

🔒 The Local-First Advantage

Modern preparation tools demand sign-ups, track analytics, rely on internet connectivity, and force you to navigate distracting web interfaces. Dooma flips this model.

• Instant Discovery: Sub-millisecond RapidFuzz search across 3,310 questions.
• Company-Focused: Access curated, up-to-date question pools for 662 companies (including Google, Amazon, Meta).
• Zero Distractions: Launch timed mock interview sessions directly in your CLI. No browser tabs required.
• Absolute Privacy: Your activity, notes, and solve rates are written directly to your local ~/.dooma/state.db. Nothing is ever uploaded.

---

🛠️ Workflows

Launch the interactive workspace simply by running dooma.

Command	Purpose
dooma practice	Interactively browse the question database
dooma browse companies	Explore question pools mapped to specific companies
dooma search <query>	Fuzzy search questions by keyword or topic
dooma sheet blind-75	Work through curated industry-standard roadmaps
dooma mock	Start a timed, randomized mock interview session
dooma dashboard	Review your solve rates, local progress, and streaks
dooma bookmarks	Access your saved and starred questions

Tip

Command Chaining: Try dooma mock --count 3 --difficulty hard to jump straight into a tough session.

Interactive Hotkeys

While inside any question flow, use these hotkeys to update your local state instantly:

Key	Action
o	Open the problem URL in your browser
m	Cycle status (unsolved → attempted → solved → skipped)
Space	Toggle bookmark
n	Write or edit a private local note
q	Return to the previous menu

---

🏗️ Architecture

Dooma is designed to be extremely lightweight, utilizing a precompiled dataset to achieve sub-second startup times.

View System Workflow

graph TD
    CLI["CLI Input"] -->|"Typer"| Fuzz["RapidFuzz Search & Filter"]
    Fuzz -->|"Reads O(1)"| Index[("Local Dataset: index.json")]
    Fuzz -->|"Cross-references"| State[("Local SQLite: state.db")]
    Index --> UI["Rich Terminal UI"]
    State --> UI
    UI -->|"'o' shortcut"| Browser["Web Browser"]

Loading

View Codebase Structure

dooma/
├── dooma/               # Core application logic
│   ├── cli/             # Typer command definitions
│   ├── dataset/         # Local-first JSON data models
│   ├── db.py            # SQLite state management
│   ├── display.py       # Rich UI rendering
│   └── loader.py        # Dataset index loading
├── scripts/             # Internal utilities (YAML -> JSON compilation)
└── tests/               # Pytest suite

---

🚀 Project Vision

Our goal is to build the definitive local-first hub for technical interview preparation. We aim to bridge the gap between problem discovery (knowing what to solve) and deliberate practice (tracking how you solve it)—all from the terminal where engineers already feel most at home.

---

🗺️ Roadmap

We are actively building the future of local-first interview prep.

Phase	Focus	Impact
Now	Pattern Taxonomy	Mapping 3,310 questions to 25 distinct patterns to enable pattern-based learning.
Now	Dataset Validation	Hardening the YAML validation pipeline to ensure high data integrity for new contributions.
Next	Curated Sheets	Integrating the complete NeetCode 150 and Striver SDE roadmaps.
Future	Spaced Repetition	Introducing an algorithm to resurface challenging questions based on local metrics.

---

🤝 Contributing

We want to make contributing to Dooma as seamless as using it. First-time contributors are explicitly welcome!

Where We Need Help

• 🐍 Python Developers: Optimize RapidFuzz search ranking, add regression tests, and expand CLI features.
• 🎨 UI Designers: Refine the Rich terminal rendering for edge cases in narrow or non-standard terminals.
• 📊 Data Enthusiasts: Help map the remaining Blind 75, NeetCode 150, and Striver SDE questions. Check issues labeled area:data.

Good First Issues

Looking for an easy place to start?

1. Check the good first issue label.
2. The highest impact task right now is adding pattern tags to existing questions in the YAML dataset.

Development Workflow

1. Clone and install dependencies:

   git clone https://github.com/im-anishraj/dooma.git
   cd dooma
   pip install -e ".[dev]"

2. Run tests and verify standards:

   ruff check .
   mypy dooma
   python -m pytest

3. Rebuild index (if modifying dataset YAMLs):

   python scripts/build_index.py

Please read our CONTRIBUTING.md guide before opening a pull request.

---

💬 Community

• GitHub Discussions - Ask questions, share ideas, and follow announcements.
• Discord - (Coming soon)

---

❓ FAQ

Does Dooma submit code for me or require a LeetCode Premium account?
No. Dooma provides metadata, problem categorization, and direct links. You do not need a premium account to use Dooma, but you will need an account on the respective platform (like LeetCode) to submit your actual code.

Where is my data stored? Can I sync it?
All state (bookmarks, notes, progress) is stored locally in ~/.dooma/state.db. You can sync this SQLite file across machines using Dropbox, iCloud, or Syncthing.

How do I completely reset my progress?
You can reset your onboarding config via dooma config --reset or safely delete the ~/.dooma directory to start completely fresh.

---

📄 License & Conduct

Released under the MIT License.
By participating in this project, you agree to abide by our Code of Conduct.