How to Build a Portfolio for Jobs: Data and Software Roles

Last updated: 21 September 2026 · By the Asuraa Team

How do you build a portfolio for jobs?

Build a portfolio by choosing two or three projects that match one target role, writing a clear README for each, and publishing them where a recruiter can open them in one click. Put the link at the top of your resume.

Quality matters more than volume. A recruiter with limited time will look at one or two projects, so make those easy to understand and easy to run.

What is a portfolio, and what should it prove?

A portfolio is a small, public collection of work that shows you can do the job you are applying for. It exists to replace claims on your resume with evidence a reader can check.

Use this two-minute test on every project. Can a stranger tell what problem it solves, what you did and what happened, and can they run or read it without asking you anything? If not, improve the README before you build anything new.

What should a portfolio contain for data and software roles?

Choose proof that matches the role. The table is editorial guidance, so adjust it to the postings you are targeting.

For project ideas by role, see our lists of data analyst portfolio projects, data scientist projects and software engineer projects for freshers.

What goes in a project README?

A README is the front page of your project. GitHub's documentation on READMEs says they typically explain what the project does, why it is useful, how users can get started, where users can get help and who maintains it.

It also says GitHub shows a README placed in the repository root, the docs directory or a hidden .github directory. A README in a public repository named after your username becomes a profile README on your profile page.

Here is a structure you can copy. This is an illustration of a data project page, not a real project.

- **Project:** Job openings by city, an analysis
- **Question:** Which cities list the most openings for data analysts in my sample?
- **Data:** 200 postings I collected on 10 Sept 2026 (CSV in /data, source noted below).
- **Method:** Cleaned titles and cities in Python, counted openings per city with SQL.
- **Result:** Two cities had the most openings; chart in /images. Limits: small, non-random sample.
- **Run it:** `python -m venv env`, install requirements.txt, run `python analyse.py`.
- **Next steps:** Add salary ranges once a reliable source is found.

Notice the limits line. Stating what your project cannot show builds more trust than claiming it proves everything.

How do you publish a portfolio site for free?

GitHub Pages is one option. GitHub's documentation on what GitHub Pages is says it is a static site hosting service that takes HTML, CSS and JavaScript files straight from a repository, optionally runs a build process, and publishes a website.

The GitHub Pages quickstart gives these steps for a user site. Follow them in order.

1. Create a repository named username.github.io, with your GitHub username in place of username.
2. Set the repository visibility and add a README.
3. Open the repository's Settings, then Pages.
4. Under source, choose "Deploy from a branch" and pick a branch.
5. Edit the README or add an index page with your content.
6. Visit username.github.io. The quickstart says changes can take up to 10 minutes to publish.

A site is optional. A tidy GitHub profile with two well-documented repositories can do the same job, and it is easier to maintain.

How do you choose which projects to show?

Pick projects that match the role you want, and show range only where it helps. Two data projects and one small software project can suit a data analyst applying to product teams. Three unrelated projects suit no one.

Then order them by how well they pass the two-minute test, with the strongest first. Pin those repositories on your profile if the option appears in your account.

What should you never put in a public portfolio?

Do not publish anything you are not allowed to share. These are editorial cautions, not legal advice.

• Confidential work data. Use public or invented data, and mention that it is invented.
• Personal data about other people. Names, phone numbers and emails in a dataset should be removed or replaced.
• Secret keys and passwords. Keep them out of the repository, including its history.
• Claims you cannot support. If your data is a sample, say so, and describe limits plainly.

How do you connect the portfolio to your resume?

Link it where a recruiter will look first. Put your profile or site URL under your name, and add one project link next to each relevant project entry.

Our guide to listing projects on a resume shows how to write the project lines. Then check how the whole resume reads with the AI resume reviewer on asuraa.in, which takes your resume, a target job title and a job description.

What do most guides on how to build a portfolio get wrong?

Many guides focus on design. These are the gaps that matter more to a recruiter.

• They start with a template, not a project. A polished site with thin projects is easy to spot, so build the projects first.
• They list ideas without evidence. Each project needs a question, a method, a result and limits.
• They forget to make it runnable. A reviewer who cannot run your code or read your data source cannot check anything.
• They pile up tutorial clones. Projects copied from a tutorial show the tutorial, so change the data, extend it and explain your choices.

FAQ

How many projects should a portfolio have?

Two or three well-documented projects are enough for most early-career applications. Choose ones that match your target role and pass a two-minute test: a stranger can see the problem, method and result. Add more only when they add different, relevant evidence.

Do I need a website to build a portfolio?

No. A GitHub profile with well-documented repositories can work. If you want a site, GitHub Pages is a static site hosting service that publishes HTML, CSS and JavaScript from a repository, and its quickstart shows the steps for a user site.

What should I put in a portfolio README?

Explain what the project does, why it is useful, how to get started, where to get help and who maintains it, as GitHub's docs describe. For a job portfolio, add the question, data source, method, result and limits so a recruiter can check quickly.

Can freshers build a portfolio without work experience?

Yes, because projects can show your skills where work history cannot. Use public or invented data, document your method and state the limits. Two solid projects with clear READMEs give a recruiter something to check.

What is GitHub Pages and is it free?

GitHub describes GitHub Pages as a static site hosting service that publishes a website from files in a repository. We did not find pricing on the pages we read, so check GitHub's current terms and plan details before relying on it.

Should a data analyst portfolio look different from a software developer's?

Yes, in what it shows. A data portfolio should show questions, cleaned data, queries or charts and written findings, while a software portfolio should show a working app or API, tests, setup steps and a readable history. Both need a clear README.

Final thoughts

A portfolio is proof a stranger can check quickly: two or three projects, clear READMEs, public or invented data and stated limits. Publish on GitHub, link it on your resume, and improve one project at a time.

If you want help with the resume side, run it through the AI resume reviewer above and browse Python and SQL learning paths to plan the skills behind your next project.