Track your daily coding sessions, set weekly goals, and get AI-generated weekly recaps of your progress.
- JWT Authentication — secure register/login with bcrypt password hashing
- Session Logging — log language, hours, what you built, and your mood
- Activity Heatmap — custom SVG heatmap of your last 16 weeks
- Weekly Goals — set coding hour targets and track progress
- AI Weekly Recap — Gemini-powered summary of your week's sessions
- Public Profiles — shareable profile URL, viewable without login
- Full Test Suite — Pytest for backend, Vitest for frontend
- CI/CD — GitHub Actions runs tests on every push
- Dockerized — Docker Compose for local development
Backend: Python 3.12, FastAPI, PostgreSQL, SQLAlchemy, Alembic, JWT (python-jose), bcrypt/passlib, Pytest, Docker
Frontend: React 19, TypeScript, Vite, Tailwind CSS, TanStack Query, React Router, Recharts, Vitest
AI: Google Gemini API for weekly recap generation
Infrastructure: Docker Compose (dev), Railway (backend), Vercel (frontend), GitHub Actions (CI)
- Python 3.12
- Node.js 18+
- Docker Desktop
git clone https://github.com/Raghavtripathii/devlog
cd devlogStart the database:
docker-compose up -dBackend:
cd backend
python -m venv venv
venv\Scripts\Activate # Windows
pip install -r requirements.txt
cp .env.example .env # fill in your values
alembic upgrade head
python -m uvicorn app.main:app --reloadFrontend:
cd frontend
npm install
npm run devBackend:
cd backend && pytest tests/ -vFrontend:
cd frontend && npx vitest runGitHub Actions runs on every push to main and every pull request into it:
- Backend job — installs dependencies and runs Pytest against a SQLite database (no Docker needed in CI)
- Frontend job — installs dependencies and runs the Vitest suite
Why FastAPI?
FastAPI generates OpenAPI docs automatically at /docs with zero extra work. Every endpoint is documented and testable without writing a single line of docs code. Pydantic v2 validation also catches bad request bodies before they reach the route logic.
Why SQLite for tests, PostgreSQL for development? Tests run against SQLite so CI doesn't need Docker or a live database. The same SQLAlchemy ORM interface means the same query logic is being tested either way. Each test gets a fresh database that's torn down after, giving full test isolation.
Why TanStack Query?
It handles caching, background refetching, loading/error states, and cache invalidation. queryClient.invalidateQueries(["sessions"]) after posting a session is one line that would take 30 lines of useEffect boilerplate — and it avoids the race conditions that come with hand-rolled fetching.
Why JWT in the Authorization header instead of cookies? The API is consumed by a separate frontend domain. Cookies would need CORS credential and SameSite handling; Authorization headers work across domains by default and are a more natural fit for a REST API that might later serve a mobile client too.
Why a custom SVG heatmap instead of a charting library? A library adds 50KB+ for something that's essentially a 2D array of colored rectangles. The custom implementation is a fraction of that size, renders faster, and gives full control over the color ramp and tooltip behavior.
devlog/
├── backend/
│ ├── app/
│ │ ├── models/ # SQLAlchemy models (user, session, goal)
│ │ ├── routers/ # API routes (auth, sessions, goals, profile)
│ │ ├── schemas/ # Pydantic request/response schemas
│ │ ├── services/ # auth, AI recap, dependencies
│ │ └── main.py
│ ├── tests/ # Pytest suite
│ ├── alembic/ # DB migrations
│ └── Dockerfile
├── frontend/
│ └── src/
│ ├── pages/ # Login, Register, Dashboard, PublicProfile
│ ├── components/ # Heatmap, SessionForm, GoalCard, etc.
│ ├── context/ # AuthContext
│ ├── lib/ # API client
│ └── types/ # Shared TypeScript types
└── .github/workflows/ # CI pipeline
Raghvendra Tripathi GitHub: @Raghavtripathii