Mini projects (course requirements or self-driven) are where skills actually consolidate — small enough to finish, real enough to teach. The rule: 2-4 weekends, one new skill each, shipped to GitHub with a README. Ideas by track:
Web track (pairs with the roadmap)
- Expense tracker with charts + localStorage
- Quiz app pulling from a question API — JS course territory
- Markdown notes app with search
- College events board (CRUD + auth) — your first full-stack loop
- Weather + air-quality dashboard for your city
Python track
- Attendance calculator (“how many classes can I skip?” — every student’s favourite math)
- PDF toolkit — merge/split/watermark CLI
- Timetable-clash checker from CSVs
- Web scraper for results/news headlines with alerts (respect robots.txt)
- Telegram bot for your class — reminders, links, polls
Data/AI-lite track
- Marks dataset analysis — trends + visualisations (analyst starter)
- Simple recommender (books/movies) with honest evaluation
- Chat-with-one-PDF using an LLM API (RAG-lite)
- Sentiment tally on public reviews of something local
- GPA predictor — regression on anonymised seniors’ data (with the caveats stated!)
Hardware-lite track (ECE/EEE-friendly)
- Ultrasonic parking/distance alert
- Temperature logger → CSV → chart pipeline
- Line-follower or obstacle bot (the classic, done cleanly)
- IR remote-controlled appliance switch (safe, isolated)
- Solar phone-charger bench test with measured curves
Extracting the full value
Finish > feature-creep; write the README as if a stranger will run it (they might — an interviewer); log what broke and how you fixed it (that log becomes interview and viva material); and let each mini feed the next — the events board becomes the final-year platform, the PDF-chat becomes the RAG assistant. Minis are how portfolios compound quietly across four semesters.
Frequently Asked Questions
What is a good mini project for second year?
Anything scoped to 2-4 weekends that teaches one new skill and ships complete — an expense tracker, attendance calculator, quiz app or sensor logger. Finished-and-documented beats ambitious-and-abandoned for both grades and GitHub.
Do mini projects matter for placements?
Cumulatively, yes — they build the skills and the GitHub trail that final-year interviews draw on, and early minis compound into stronger major projects. Interviewers read consistent small builds as genuine capability.
Should mini projects use AI?
One AI-flavoured mini (a chat-with-PDF tool, a simple recommender with real evaluation) is excellent 2026 practice — after fundamentals-first minis. Use AI APIs as components you understand, not as magic you can’t explain in the viva.