A computational design portfolio checklist comes down to what a recruiter can confirm in about 30 seconds: your target role, the brief, the rule or script behind the form, a few honest iterations, a measurable outcome, and what you personally did.
Most portfolios from Indian students and early-career designers don't fail because the work is weak. They fail because the process is invisible. Glossy renders sit up front and the Grasshopper definition is nowhere.
This guide gives you the checklist, breaks one project down against it, and shows the mistakes that get a portfolio skipped. Run your own through it before you send it anywhere.
Our other guides cover how to build the portfolio from scratch first. This one checks whether it's good enough to send.
What a Recruiter Actually Looks At First
In roughly the first 30 seconds, a recruiter checks four things: role clarity, project relevance, proof of process, and one number. That 30-second window is a rule of thumb. Hiring managers who write publicly about how they screen keep landing on about the same figure. It isn't a formal study, but it matches how this goes in practice.
Seconds 1–5: do you say what you are? Your headline should read "Computational Designer" or "Parametric Modeller," not "Passionate architect who loves design." If someone has to guess your specialisation, they move on.
Seconds 5–10: do the project titles mean something? "Parametric facade panelisation for a west-facing office" tells them more than "Flow" or "Echoes." Cryptic names might feel artsy, but they cost you the click.
Seconds 10–20: can they see how you built it? A node cluster, a script snippet or a few iteration frames beats another hero render. Process over polish.
Seconds 20–30: is there proof it worked? A number: options generated, panels reduced, material saved, hours cut.
Competition is real here. The Council of Architecture says about 370 institutions in India currently offer recognised architecture programmes. A lot of those graduates apply to the same studios and design-tech teams, so a portfolio that explains itself fast has a genuine edge.
Also Read: Computational Design Resume Guide
Want projects that pass this checklist? Kaarwan's Rhino and Grasshopper course helps you build real definitions, iterations and outputs you can proudly put in your portfolio.
The Checklist: Brief, Rule/Script, Iterations, Outcome, Your Contribution
Every project in your portfolio should clear these five checks, and the table below is the quick version.
| Check | What the recruiter wants to see | Quick self-test |
| Brief | The problem, site or constraint in two lines | Can a stranger restate the brief after one read? |
| Rule/Script | The logic: a definition screenshot, key parameters, script level | Could someone rebuild your idea from this page? |
| Iterations | A handful of options, including what you rejected | Is there more than one version on show? |
| Outcome | A measurable result or a clear fabrication/output file | Is there at least one number? |
| Your contribution | What you did versus the team or tutor | Does "I" appear in the first line of each section? |
Start with the problem, not the render. Two lines is enough: what the project needed and what constrained it. Without it, the rest has no anchor.
Rule/Script: This is the part recruiters scan for. Show the definition (a clean screenshot, grouped and labelled), name the three or four parameters that drive the design, and say honestly how deep your scripting goes. "GH Python (basic)" beats an overclaim that falls apart in a live interview. If you're unsure which skills to flag, read which Grasshopper skills employers look for.
Iterations: Show three to six options in a tidy strip, then circle the one you picked and say why. Showing a rejected version is a strength. It proves the model responded to your decisions.
Outcome: Give a number, even a small one. "Generated 24 facade options, shortlisted 3 on shading performance" is a result. If it reached a laser cutter, a CNC file or a 1:50 model, show that too. Digital fabrication output is proof the parameters connected to something physical.
Your contribution: Group project? Say what you built. A reviewer can't give you credit for a decision they can't find.
Skipped alert: a portfolio with five beautiful projects and zero visible logic is usually skipped by someone who never doubted your talent. They simply had nothing to verify.
On quantity, aim for three to five solid, end-to-end projects. Not every script you've ever written.
Parametric Design Portfolio Examples: One Project Broken Down
Let's walk one project through the checklist. This is an illustrative example (the numbers are sample numbers, not real results), so swap in your own.
Imagine a final-year studio project: a perforated screen for a west-facing office facade in Bengaluru, where afternoon glare is the real problem.
Brief: "Reduce afternoon glare on a west facade without killing the view." Two lines, done.
Rule/Script: One definition. Hole size and spacing driven by the sun angle on each panel, with a Ladybug radiation analysis feeding back into it. Screenshot grouped into four labelled clusters: surface, analysis, aperture logic, output.
Iterations: A strip of six options, with the circled choice and a one-line reason, like "Option 4 cut direct radiation most without closing the view."
Outcome: Say you generated 24 options and shortlisted 3, then sent the final panel set to a laser cutter for a model. Two numbers, one physical output.
Your contribution: "I wrote the definition and ran the analysis. A teammate handled the render."
Nothing here needs a big firm name behind it. Studio work, your thesis or work from your practical training semester can all carry a project like this. For more on structure, our general portfolio-building tips are a good companion.
Also Explore: Parametric Design Career Roadmap
Common Mistakes That Get a Portfolio Skipped
These come up again and again.
- Renders without logic. Final images only, no definition, no iterations.
- A vague headline. "Architect | Designer | Creative" tells a computational hiring team nothing.
- Overclaiming scripting. Listing "Python (advanced)" after two scripts is a trap you'll walk into in the interview.
- A portfolio nobody can open. Check your Google Drive link is set to "Anyone with the link can view." Then open it on your phone, on mobile data. If it crawls, assume the recruiter closes it. If you host on Behance, here's hosting your portfolio on Behance done properly.
- A dump of everything. Thirty projects with no explanation reads as a lack of judgement, not range.
And please link the portfolio from your resume. They work as a pair.
Also Read: Computational Design for Architects: What It Actually Means
How This Differs From an Architecture Portfolio
A general architecture portfolio sells the final image: the concept, the drawings, the model. A computational design portfolio sells the rule behind the image. A conventional reviewer asks, "Is this a good building?" A computational reviewer asks, "Did this designer build a system, and can they explain it?" So you need less hero imagery and more definition screenshots, parameter logic and option strips.
You can keep your beautiful renders. They just can't be the only evidence. If you're aiming at computational design jobs, this shift in emphasis is the biggest single upgrade you can make.
Portfolio feeling thin on process? Learn parametric workflows hands-on with Kaarwan's Rhino and Grasshopper program, and turn every project into proof recruiters can verify.
Three Takeaways
- Make the logic visible: definition, parameters and honest scripting level.
- Give every project a brief, a few iterations and at least one number.
- Keep it to three to five projects you can explain out loud.
If your project evidence is thin, building it is the real fix. Our Rhino and Grasshopper course is designed around exactly that kind of project work.
FAQs
What should a computational design portfolio include?
Three to five end-to-end projects, each with a brief, a clear view of the rule or script, a few iterations, an outcome with at least one number, and your specific contribution. Add a headline that names your specialisation and a plain-text link from your resume.
How many projects should be in a parametric design portfolio?
Three to five is a good range. Depth beats volume. One project with visible logic and a measured result does more than ten pretty ones with no explanation.
Do recruiters care about the script or just the final render?
They care about both, but the script is what separates you from other applicants. A render shows the result. The definition shows you can repeat and adjust the result, which is the job.
Should I include failed iterations in my portfolio?
Yes, one or two. A rejected option with a short reason shows judgement and proves your model responded to decisions. Just don't bury the final choice. Circle it and say why you picked it.
