“I should build a side project” is easy to say and hard to do when you are juggling a return to tech, daily life, and self-doubt about coding skills. The hardest part is picking something achievable. This post is a list of concrete project ideas, each scoped to a specific time budget, with the technical skills they showcase.
Weekend projects (10–20 hours)
1. Personal CLI dashboard
A terminal command that fetches your day: weather, calendar events, RSS feeds, GitHub PRs needing review.
Skills: CLI design, async/await, API integration, JSON parsing.
2. Slack reminder bot
A bot that posts daily standup prompts to a Slack channel.
Skills: webhooks, scheduling, OAuth, basic backend.
3. URL preview service
A small API that returns a screenshot + title + description for any URL.
Skills: headless browsers (Puppeteer), HTTP, caching.
One-week projects (40–60 hours)
4. Personal expense tracker
Web app to log expenses, categorize, and visualize. Authentication, persistence, charts.
Skills: full-stack, basic auth, database modeling, charts.
5. Recipe scaler
Given a URL or pasted recipe, scale all ingredients to a target servings count. Bonus: dietary substitutions.
Skills: text parsing, NLP basics, web scraping (or LLM call).
6. Markdown notes app with sync
Local-first markdown editor that syncs to Dropbox or your own server.
Skills: conflict resolution, offline-first, sync protocols, local storage.
Two-week projects (60–100 hours)
7. URL shortener with analytics
Like bit.ly. Custom domains, click analytics, geo-distribution stats.
Skills: backend, Redis, geo IP, traffic analytics, dashboards.
8. Headless feature flag service
A self-hostable feature flag API. Boolean and percentage rollout flags.
Skills: backend, SDK design, real-time updates, multi-tenant.
9. RSS reader with categories
Modern RSS reader. Categorize feeds, mark as read, search.
Skills: RSS parsing, full-text search, sync, UX.
Month-long projects (100–200 hours)
10. SaaS product with payments
Identify a real (small) need; build a SaaS that solves it. Stripe integration. Even if you only get 5 users, this is the strongest portfolio item.
Skills: end-to-end product engineering, payments, deployment, marketing.
11. Mobile app on the store
A small but useful native app. Could be a habit tracker, a hobby tool, or a niche utility.
Skills: Swift/Kotlin, App Store submission, iCloud/Firebase sync.
12. ML-powered tool
Build something using an LLM API or a small open-source model. Examples: writing assistant, summarizer, image classifier.
Skills: ML basics, prompt engineering, latency management, deployment.
The “I want to learn X” project
If your goal is learning a specific stack:
- Rust: a small CLI tool that processes structured data
- Go: a microservice with health checks and graceful shutdown
- TypeScript: rewrite an existing JS project of yours in TS
- React Native: port a small web app you built to mobile
- Kubernetes: deploy any of the above to a real cluster
Project completion is more important than ambition
A finished URL shortener beats a half-finished Twitter clone every time. Pick something you can ship in your available time.
Frequently Asked Questions
What if I get bored mid-project?
Either ship a smaller version (call it v1, write it up) or move on. Half-finished projects in public are negative signal.
Should I open-source my projects?
Yes. Public GitHub repos are easier for recruiters to find. Even if no one stars them, they exist as evidence.
How do I avoid over-engineering?
Set a deadline, write a one-page spec at the start, and resist refactoring until v1 ships.