6 AI Portfolio Projects for Instructional Designers
By Devlin Peck · Updated
Part of the Instructional Design Portfolios guide
The best AI portfolio projects prove that you can design learning experiences with AI and incorporate AI-powered practice, not just generate content faster. This guide gives you six projects you can actually build, ranked easiest first, from a weekend custom GPT to an AI practice simulation embedded in a Storyline 360 flagship. For each one, you'll see what it proves to a hiring manager, the tools involved, a realistic timeline, and how to present it.
This piece extends my full guide to creating an instructional design portfolio. That article covers the portfolio itself: structure, write-ups, and how to market it. This one answers the narrower question I hear constantly now: what should the AI projects on that portfolio actually be?
Why do instructional design portfolios need AI projects?
Because hiring managers now treat AI fluency as a screening requirement, and your portfolio is the only place you can prove it rather than just tell people about it. Anyone can list "AI tools" on a resume. A working project that a hiring manager can click, experience, and evaluate is a whole other level of evidence.
The hiring data backs this up:
| Finding | Source |
|---|---|
| 66% of leaders say they would not hire someone without AI skills | Microsoft & LinkedIn 2024 Work Trend Index |
| 71% of leaders would rather hire a less experienced candidate with AI skills than a more experienced one without them | Microsoft & LinkedIn 2024 Work Trend Index |
| AI literacy is the #1 fastest-growing skill in the US | LinkedIn Skills on the Rise 2025 |
| Training and development specialist employment is projected to grow 11% from 2024 to 2034 | BLS Occupational Outlook Handbook |
The field is growing, but the candidates winning the interviews look different than they did a few years ago. The shift I'm seeing in 2026 is that the strongest portfolios no longer treat Storyline 360 skills and AI readiness as separate tracks. They combine traditional eLearning development and modern AI capabilities in every project.
The strongest instructional design portfolios in 2026 don't separate Storyline 360 skills from AI readiness.
I've also seen this play out on the job-search side. Applying to roles with your portfolio prominently linked goes a long way in this hiring market, and it goes further when the portfolio includes AI-forward projects. LinkedIn, Indeed, and referrals are still where most opportunities come from, but the portfolio is what converts your applications into interviews.
What makes a strong AI portfolio project?
A strong AI project demonstrates judgment and tradeoffs on a real problem. It shows an AI-powered learning experience rather than just AI-generated content. This distinction is the single most important idea in this article.
AI-generated content means you used AI to produce assets: a script, images, a voiceover. It's useful, but every candidate can do it, so it proves very little on its own. An AI-powered learning experience means AI is part of how the learner learns: a conversation partner that responds to what they say, a coach that gives feedback on their choices, or a simulation that adapts. Designing one of those requires actual instructional design thinking, and hiring managers can tell the difference in about ninety seconds.
Here's what an AI-powered learning experience feels like from the learner's side. Play the detective simulation below: you interview suspects to solve an art heist, and the conversation responds to whatever you ask.
When I talk about standing out as an instructional designer in 2026, I keep coming back to five things:
- Build projects that show judgment, tradeoffs, and problem-solving. Don't just add polish for its own sake.
- Use instructional design theory to justify decisions. Don't name-drop models; show why you made the call.
- Apply AI to accelerate real workflows. Analysis, iteration, and prototyping.
- Design AI-powered learning experiences, not just AI-generated content.
- Show comfort with ambiguity and business constraints. Real projects have budgets, deadlines, and stakeholders.
Every project in the roster below is designed to hit several of these at once. And whichever you build, it needs a short case study alongside it: the problem, the audience, the decisions you made, and what you cut. This write-up discipline is what separates a portfolio piece from a file dump.
The 6 best AI portfolio projects to build
Ranked easiest first: a custom GPT, an AI practice simulation inside a Storyline 360 flagship, an AI chatbot coded into Storyline with JavaScript, an AI-assisted design workflow case study, and a scenario course with AI-generated media. And after all of that: your portfolio site rebuilt with AI. Here's how they compare:
| Project | What it proves | Tools | Realistic timeline | Coding required |
|---|---|---|---|---|
| 1. Custom GPT | AI fluency & problem-solving for a real audience | ChatGPT (GPT builder) | A weekend | No |
| 2. AI practice simulation in a Storyline 360 flagship | Scenario design, AI integration, and traditional eLearning development together | Storyline 360, devlin.ai | 2 to 6 weeks | No |
| 3. AI chatbot in Storyline with JavaScript | Technical development skill | Storyline 360, JavaScript, an AI API | 4 to 8 weeks | Yes (JavaScript) |
| 4. AI-assisted design workflow case study | Judgment across analysis, iteration, and prototyping | Claude Code & your design tools | 2 to 6 weeks | No |
| 5. Scenario course with AI-generated media | Production range and media judgment | Synthesia, ElevenLabs, an authoring tool | 2 to 3 weeks | No |
| 6. Portfolio site rebuilt with AI | Modern workflow plus a better web presence | Claude Design, devlin.host | A weekend | No |
1. A custom GPT that solves a real problem
Since mid-2025 I've been arguing that a custom GPT is the single best first AI project for instructional designers. It shows AI skill, it solves a real problem for a real audience, and it stands out far more than yet another generic eLearning course.
A custom GPT is a version of ChatGPT you configure with instructions, knowledge files, and a defined role. The design work is pure instructional design: who is the learner, what do they struggle with, what should the assistant do (and refuse to do), and how do you know it works?
The proof this works: a viewer who watched my video recommending this project built a custom GPT, put the AI skillset on their portfolio site, and landed a part-time role specifically as an AI-powered instructional designer.
For a concrete example of what this looks like at a high level, one of our students, Catherine Schwebel, built an AI-powered simulation using ChatGPT that helps nurses, physician assistants, and discharge educators practice delivering clear, empathetic post-op wound care instructions. Check out her write-up linked above to see where the bar is.
Tools: ChatGPT with the GPT builder. OpenAI's Creating a GPT help page covers the mechanics; the instructional design is on you.
How to present it: link the live GPT so a hiring manager can try it, and pair it with a short case study covering the audience, the problem, the instructions and knowledge you gave it, and a sample transcript showing it working.
2. An AI practice simulation inside a Storyline 360 flagship
This is the strongest single project on the list, and it's the one I'd point at if you have Storyline access and a few weeks to invest. It combines the two things hiring managers most want to see in one artifact: traditional eLearning development in Storyline 360 and a modern AI practice layer where the learner holds a real conversation.
Here's how this project works: build a scenario-based Storyline course around a skill people actually struggle with (difficult feedback conversations, patient communication, sales objections), then embed an AI conversation simulation where the learner practices the skill instead of clicking through multiple-choice approximations of it.
Full disclosure: devlin.ai is my company, and this is exactly the problem it exists to solve. You describe the scenario in plain English and get a working text or voice simulation that you can embed in Storyline, Rise, or any LMS. A Storyline 360 variable bridge lets the sim update Storyline variables like score and pass/fail, so your course can branch based on how the conversation went. I also created a Starter plan for aspiring IDs and portfolio-builders, and it runs $9/month ($6/month paid annually) as of August 2026.
Describe a scenario your learners need to practice and try the simulation it builds, right here from this article (this is an example of what would be embedded in your Storyline course, with the look-and-feel customized to match your project):
Scope: this is a flagship, not a weekend build. Plan on 2 to 6 weeks depending on your Storyline experience. Pair it with a case study on the scenario design itself: why this skill, how you structured the difficulty, and how you defined what "good" looks like in the conversation.
3. An AI chatbot built into Storyline with JavaScript
If you want to prove technical development skill, wire an AI model into Storyline yourself with JavaScript. My tutorial on how to build an AI chatbot in Articulate Storyline walks through the build path.
One of our Peck Academy graduates, Daniel Guimont, did this well: he used JavaScript to incorporate AI models directly into Storyline 360 to power an AI character inside the course.
Be clear-eyed about the tradeoff, and say so in your write-up. Hand-coding means more control and a stronger technical story, but you own the API-key handling and you don't get the evaluation, scoring, and reporting a purpose-built practice layer gives you. Documenting how you handled the API key and why you designed the prompt the way you did turns the limitations into evidence of judgment, which is exactly what the write-up is for.
Scope: 1 to 2 weeks if you're comfortable with basic JavaScript, 4-8 weeks if you're not. This is the only project on the list that requires code.
4. An AI-assisted design workflow case study
This project proves that you can use AI to accelerate real instructional design work: analysis, iteration, and prototyping. The artifact is the documented process itself (plus supporting screenshots).
Take one design problem from brief to prototype and show your work at every step: the prompts you used for needs analysis, what the AI produced, what you kept, what you overrode, and why. A "what the AI got wrong" section is the most persuasive part of the whole piece, because it proves the judgment is coming from you.
This is also the project that answers the question aspiring IDs ask me most about AI portfolios: how much AI-generated material is acceptable? This project shows that you can include any amount, as long as the decisions are visibly yours and documented.
To take this a step further, you can build an automated pipeline that takes raw inputs and turns them into useful outputs; for example, you can use a tool like Claude Code to create a pipeline that takes raw SME notes and screenshots, and then turns it into a fully functional software simulation prototype. More advanced, but a bigger payoff (both for your actual work and for your portfolio).
Tools: whatever you actually use. My roundup of AI tools for instructional design covers the current stack if you're building yours.
Scope: 1 to 2 weeks if not building an automated pipeline, 2 to 6 weeks if you are. If doing the former, this is mostly writing and screenshotting the work that you'd be doing anyway.
5. A scenario-based course with AI-generated media
This one shows production range: a scenario-based course that uses AI avatar video (for example, Synthesia), AI voiceover (for example, ElevenLabs), and AI-generated imagery. The case study explains where AI media was the right call, where it wasn't, and how you kept the quality bar high.
Scope: 2 to 3 weeks. Keep the course short and the scenario sharp.
6. Rebuild your portfolio site itself with AI
Here's the meta-project: use an AI design workflow to build the portfolio website that holds everything else. It proves a modern workflow and upgrades your web presence in the same weekend. Sabrina's full tutorial covers the exact build, from design through exporting and hosting:
For hosting, devlin.host is free HTML project hosting inside devlin.ai, and it can cover a full portfolio site along with projects. The Starter tier adds a custom domain if you want yourname.com, and this gives you enough credits to include AI simulations in your portfolio, too.
Scope: a weekend. Don't let the site itself eat the time your projects need.
How should you present AI projects in your portfolio?
Every AI project needs a live, clickable artifact plus a case study explaining the problem, your design decisions, and why your solution was the right call. A screenshot gallery is not a portfolio piece. If a hiring manager can't experience the thing working, you've thrown away the main advantage AI projects have: they're demonstrable.
Structure each write-up the same way:
- The problem and the audience. Who struggles with what, and why it's worth solving.
- Your design decisions and tradeoffs. What you chose, what you cut, and the reasoning.
- Your prompts and iterations. For AI projects, the prompt history is your design documentation. Show your iteration process.
- What you'd do differently. One paragraph of self-critique reads as senior judgment.
Then make the artifact reachable in two clicks from your homepage. For Storyline projects, my guide on how to upload your Storyline project to the web covers getting a shareable link. Keep the overall portfolio shape simple: one flagship plus a few supporting pieces. Not every supporting project needs to be AI-heavy; my list of eLearning portfolio project ideas has plenty of options, and my portfolio reviews of real ID portfolios show how finished portfolios handle write-ups and navigation. Three strong samples with context beat ten mediocre ones every time.
How do you choose which AI project to build first?
Match the project to the skill you most need to prove and the time you actually have. A finished custom GPT this weekend beats a half-finished flagship in three months, and stalled, over-scoped flagships are the most common way I see people lose momentum.
When choosing a topic, the choice should be simple: combine your domain passion with a real problem. Back in early 2025, I tested this on myself. I fed a list of my own passions into the become-an-instructional-designer AI assistant I'd built and trained on our process, and asked it for portfolio project ideas. It suggested several I had already come up with on my own. The tool was useful, but the underlying method is the point: start from what you know and care about, find where people in that world actually struggle, and design for it. AI can speed up the brainstorm, but it can't replace knowing your audience.
One aside, with my ownership disclosed: if you're a career transitioner who wants structure and feedback rather than building on your own, Peck Academy is my licensed career school, and students there build exactly this kind of AI-integrated flagship project in Storyline 360 with extensive feedback, plus a portfolio website. The curriculum is reviewed by an advisory committee of 10 independent industry experts. Everything in this article, though, is doable on your own.
Frequently asked questions
What should an instructional designer have in their portfolio?
One to three strong samples: one flagship project that shows your full instructional design process, plus 1 or 2 supporting pieces that show specific skills like storyboarding, needs analysis, or AI integration. Every sample needs a short case study explaining the problem, the audience, and your decisions. The full portfolio guide covers structure, write-ups, and marketing in depth.
Do hiring managers actually care about AI portfolio projects?
Yes. In the 2024 Work Trend Index, Microsoft and LinkedIn found that 66% of leaders would not hire someone without AI skills, and 71% would prefer a less experienced candidate with AI skills over a more experienced one without them. My own read of the current market agrees: applying with a prominently linked portfolio goes a long way, and AI-forward projects make it go further.
How many AI projects do you need?
One AI-integrated flagship project is enough to start applying, and it does get people hired. A single project combining traditional eLearning skills like Storyline 360 with modern AI skills covers the base. That said, in 2026 and beyond, the best practice is to integrate AI into every project in your portfolio, unless you know you're applying into an industry where AI adoption is still limited. And that exception is shrinking fast.
Do you need to know how to code to build AI portfolio projects?
No, not for five of the six projects here. Custom GPTs, devlin.ai simulations, workflow case studies, AI-media courses, and AI-built portfolio sites all require zero code. The one exception is the JavaScript chatbot in Storyline (project 3), which requires basic JavaScript and is worth building only if technical development is the skill you want to prove.