How to design AI interfaces that actually work for people.
AI doesn’t have to be slow, verbose and unreliable. We can shape AI experiences way beyond a chat — and guide users to better prompts and results without typing. This course helps with just that.
You’ll learn actionable design patterns for AI, real-life examples and AI research, how to lead conversations with engineers and PMs — and how to design realistic and feasible AI experiences that users trust, love and use.
What can we design with AI today? We'll study the different flavors of AI UX, the emergence of AI fatigue and how people actually use AI. We’ll also touch upon product-market-fit gap, quiet vs. visible AI, AI search optimization — and why "AI-second" products might work better than "AI-first" products.
How magical is AI really? Well, let’s take a closer look at what actually happens inside of that magical box. We’ll break down confusing AI vocabulary to have meaningful conversations with AI engineers and product managers.
Duration
Lesson number:
System & User Prompts
That’s where our UX journey begins! As designers, we can shape and influence how AI responds to user’s queries. It all starts with system and user prompts — and custom instructions written by users.
Duration
Lesson number:
Context Windows
For an intelligence of any kind, AI is remarkably forgetful — especially during longer conversations with users. It’s not surprising because it actually has a limited “memory”, or “context window”.
Duration
Lesson number:
Temperature
Well, it’s uncharted waters now! Let’s explore temperature, top_p, top_k and max_tokens — slightly mysterious but very useful AI parameters that control how “creative”, predictable and verbose AI's responses are.
Duration
Lesson number:
Memory Management
We often hear about AI having “memory”, but how does it actually work? We’ll explore how AI mimics memory, the critical UX challenges that memory management creates and why we need to set up different personas and give users control over what the AI remembers about them.
Duration
Lesson number:
Prompt Engineering
As humans, we aren’t very good at articulating our intent well. So why do most AI products place the burden of crafting the perfect prompt on users' shoulders?
Let’s see how we can change that, and why why a good old-fashioned UI with nudges and controls is so much needed to complement an empty text box.
Duration
Lesson number:
Slow AI vs. Fast AI
Over the years, AI experiences have changed. Compared to the early days, AI today spends way more time planning, researching, “reasoning”, asking for more context and stitching data together. For some tasks users need slow AI, and for others AI must deliver results faster, despite its drawbacks.
Duration
Lesson number:
AI Agents
The most significant attribute of AI agents is autonomy. But how much control are we willing to give AI agents to take over and do our work for us? We’ll look at how AI agents work, where they struggle, how they are organized and why it’s necessary to keep AI agents on a leash.
Duration
Lesson number:
AI Constraints and Limitations
AI has inherent flaws and limitations. As designers, we need to know these constraints to design viable and feasible AI experiences. We’ll explore AI naivety and hallucinations, slow speed and low accuracy, forgetfulness and AI model collapse.
Duration
Lesson number:
Digital Sustainability
No technology is free — it always has some costs attached to it. In fact, AI has an enormous environmental cost, from massive energy consumption and toxic mining to carbon emissions and water required for cooling.
Duration
Lesson number:
Usability
How do people actually use AI? How do AI experiences fit into their daily workflows? And what do we know about common user behavior patterns and shortcomings of AI experiences today?
Duration
Lesson number:
Feature Discoverability
For any AI feature to deliver value, it must first be discovered — and that’s where the troubles begin. Prompt suggestions often go unnoticed, “sparkles” are often misunderstood, and even prominent AI assistants get ignored.
Duration
Lesson number:
Capability Awareness
To many, an empty text box is remarkably scary. It suggests to “Ask me anything”, but then users don’t really know what to ask, and what format, and at which level of detail.
Let’s see how we can help users understand what AI can do for them — e.g. with task-oriented suggestions, good old-fashioned filters and UI controls.
Duration
Lesson number:
Context Awareness
Context awareness is knowing what it remembers and why it generated a specific response. Making AI’s context transparent creates trust, but it can go way beyond showing sources.
Let’s see how we can show scope with context chips, how to map statements to specific text fragments and give users better overview of what AI knows about them.
Duration
Lesson number:
Onboarding AI Features
Key AI features are often left between the lines, inviting assumptions and guesses. And typically that’s the job of onboarding.
Let’s see how we can design better onboarding for AI features, from pre-prompts and interaction modes to custom user personas and recipe guides.
Duration
Lesson number:
AI Design Framework
The time has come! Let’s dive into the myriad of practical design patterns for AI interfaces. Let’s explore how we can address them — from scaffolding and interaction to input UX, output UX, refinement and orchestration.
Duration
Lesson number:
Scaffolding (Part 1)
So where should AI actually live in your product? Let’s dive into “scaffolding” — the layout, composition and integration of our AI features.
We’ll explore center-stage experiences, collapsible widgets, side panels and complex flow builders. We’ll see how AI can go way beyond a text box.
Duration
Lesson number:
Scaffolding (Part 2)
We’ll drill down into very impactful but underutilized patterns for scaffolding. We’ll explore inline overlays, infinite canvases, AI-powered data grids, third-party systems and the rise of the AI-enhanced search.
Duration
Lesson number:
Interaction Design
We often think about chat input, but there are different interaction modes — from the familiar text and voice inputs to visual, haptic and ambient interactions.
And then, we also have multi-modal experiences, and we can even turn a simple doodle into a working prototype.
Duration
Lesson number:
AI Chat Interfaces
Designing an AI-human chat is an entirely different beast than designing a social messaging experience like WhatsApp or Telegram.
People are very good at spotting an AI disguised as a “human assistant”. We’ll explore design guidelines for transparent, useful and respectful human-AI interactions.
Duration
Lesson number:
Voice UIs
For AI features on mobile, voice is often a default expectation, as typing is tiring and slow. We’ll study UX challenges of designing invisible UIs, why it’s so critical to be intentional with voice and tone, the 10-second rule, and how to create an accessible voice experience.
Duration
Lesson number:
Conversational UIs
A good conversation is built on trust, understanding, clarity and relevance. Yet many chatbots run in loops, making users work harder, rather than simplify the experience for them.
We’ll explore how to design dialogue flows, shape voice and tone and UX guidelines for more meaningful and helpful Human-AI conversations.
Duration
Lesson number:
Designing Input UX (Part 1)
When getting answers is easy, we have to figure out how to ask the right questions. They must be detailed enough and specific enough. So let’s dive into the art of designing the input UX, modes of prompting and how to constrain the AI upfront with scoping and filtering.
Duration
Lesson number:
Designing Input UX (Part 2)
Good prompts don’t have to require a lot of typing. We can use structured prompts, templates and presets, daemons and personas that change the AI's perspective, task builders that construct prompts and visual canvases for intuitive interactions.
Duration
Lesson number:
Designing Output UX
Users are often confronted with a wall of text produced by AI — and they often get lost somewhere between sentences, especially in longer conversations. How do we make output more useful?
We'll explore the need for forced ranking, the consensus meter, color-coding, style lenses and collapsible reasoning traces to help users make sense of the AI output instantly.
Duration
Lesson number:
Accessibility
Great UX is always built on accessible, inclusive experiences — and AI is certainly no exception. AI UIs have critical accessibility challenges, from the inverted navigation nightmare in chatbots to noisy, polluting streaming.
And we’ll see why conversations about AI accessibility start not with the UI, but with the data that AI has been trained on.
Duration
Lesson number:
Refinement UX
The most frustrating part of the AI experience happens after the AI gives its first response — well, welcome to the refinement journey!
We'll explore how to make refinement less annoying and less time-consuming — with direct interactions, contextual prompts and precision knobs to get the exact result users need, faster.
Duration
Lesson number:
Orchestration UX
As AI takes over tactical tasks, our role as designers is evolving from sketching UIs to orchestrating flows. We’ll explore what it means to direct, monitor, and intervene with AI.
We’ll see why building guardrails, permissions and approval flows is so critical. This is our new job: keeping the AI on a leash, so it is always aligned with human values.
Duration
Lesson number:
Designing Guardrails and Permissions
In times when AI can act on user’s beahalf — by sending emails, spending money and initiating actions — we need to establish robust guardrails and a safety net to avoid AI agents going off the track.
Duration
Lesson number:
AI Capability / Value Matrix
Not every AI feature can deliver on its promises, and not every problem needs an AI solution. We can map out AI capabilities against the value they deliver, to prioritize, estimate and choose AI initiatives to work on. That’s usually called AI Capability/Value Matrix.
Duration
Lesson number:
Design Workflow For AI Features
Once we identified high-value opportunities with AI, we need to start designing AI flows. How exactly would it work though? Let’s explore how design teams create, iterate on and evolve AI features from scratch — from low-fidelity AI prototypes to testing, refinements and usability testing.
Duration
Lesson number:
How To Measure AI UX (with AI Evals)
How do we measure the quality of AI experiences? How do we track AI performance over time, and its impact on UX, loyalty, retention and quality of output?
Meet the world of AI Evals — a way of tracking the performance and reliability of AI systems. Let’s dive into the specifics of how we can make it work for AI features.
Duration
Lesson number:
How To Build Trust and Confidence
AI is fragile, and often mistakes aren’t an exception, but rather a matter of time. And every time a user discovers a mistake, it’s a small betrayal of trust. Mistakes are expensive as each betrayal chips away from the carefully orchestrated relationship with the user.
Duration
Lesson number:
Next Steps
You’ve made it all the way to the last session! But what’s next? Just a few closing thoughts on next steps — and how to feel comfortable and confident with the ever-evolving AI. Plus, the role of humans in that AI world, and the values that we bring to the table.
Duration
Simple prices, no surprises. Buy once, access forever.
What can we design with AI today? We'll study the different flavors of AI UX, the emergence of AI fatigue and how people actually use AI. We’ll also touch upon product-market-fit gap, quiet vs. visible AI, AI search optimization — and why "AI-second" products might work better than "AI-first" products.
How magical is AI really? Well, let’s take a closer look at what actually happens inside of that magical box. We’ll break down confusing AI vocabulary to have meaningful conversations with AI engineers and product managers.
That’s where our UX journey begins! As designers, we can shape and influence how AI responds to user’s queries. It all starts with system and user prompts — and custom instructions written by users.
For an intelligence of any kind, AI is remarkably forgetful — especially during longer conversations with users. It’s not surprising because it actually has a limited “memory”, or “context window”.
Well, it’s uncharted waters now! Let’s explore temperature, top_p, top_k and max_tokens — slightly mysterious but very useful AI parameters that control how “creative”, predictable and verbose AI's responses are.
We often hear about AI having “memory”, but how does it actually work? We’ll explore how AI mimics memory, the critical UX challenges that memory management creates and why we need to set up different personas and give users control over what the AI remembers about them.
As humans, we aren’t very good at articulating our intent well. So why do most AI products place the burden of crafting the perfect prompt on users' shoulders?
Let’s see how we can change that, and why why a good old-fashioned UI with nudges and controls is so much needed to complement an empty text box.
Over the years, AI experiences have changed. Compared to the early days, AI today spends way more time planning, researching, “reasoning”, asking for more context and stitching data together. For some tasks users need slow AI, and for others AI must deliver results faster, despite its drawbacks.
The most significant attribute of AI agents is autonomy. But how much control are we willing to give AI agents to take over and do our work for us? We’ll look at how AI agents work, where they struggle, how they are organized and why it’s necessary to keep AI agents on a leash.
AI has inherent flaws and limitations. As designers, we need to know these constraints to design viable and feasible AI experiences. We’ll explore AI naivety and hallucinations, slow speed and low accuracy, forgetfulness and AI model collapse.
No technology is free — it always has some costs attached to it. In fact, AI has an enormous environmental cost, from massive energy consumption and toxic mining to carbon emissions and water required for cooling.
How do people actually use AI? How do AI experiences fit into their daily workflows? And what do we know about common user behavior patterns and shortcomings of AI experiences today?
For any AI feature to deliver value, it must first be discovered — and that’s where the troubles begin. Prompt suggestions often go unnoticed, “sparkles” are often misunderstood, and even prominent AI assistants get ignored.
To many, an empty text box is remarkably scary. It suggests to “Ask me anything”, but then users don’t really know what to ask, and what format, and at which level of detail.
Let’s see how we can help users understand what AI can do for them — e.g. with task-oriented suggestions, good old-fashioned filters and UI controls.
Context awareness is knowing what it remembers and why it generated a specific response. Making AI’s context transparent creates trust, but it can go way beyond showing sources.
Let’s see how we can show scope with context chips, how to map statements to specific text fragments and give users better overview of what AI knows about them.
The time has come! Let’s dive into the myriad of practical design patterns for AI interfaces. Let’s explore how we can address them — from scaffolding and interaction to input UX, output UX, refinement and orchestration.
We’ll drill down into very impactful but underutilized patterns for scaffolding. We’ll explore inline overlays, infinite canvases, AI-powered data grids, third-party systems and the rise of the AI-enhanced search.
We often think about chat input, but there are different interaction modes — from the familiar text and voice inputs to visual, haptic and ambient interactions.
And then, we also have multi-modal experiences, and we can even turn a simple doodle into a working prototype.
Designing an AI-human chat is an entirely different beast than designing a social messaging experience like WhatsApp or Telegram.
People are very good at spotting an AI disguised as a “human assistant”. We’ll explore design guidelines for transparent, useful and respectful human-AI interactions.
For AI features on mobile, voice is often a default expectation, as typing is tiring and slow. We’ll study UX challenges of designing invisible UIs, why it’s so critical to be intentional with voice and tone, the 10-second rule, and how to create an accessible voice experience.
A good conversation is built on trust, understanding, clarity and relevance. Yet many chatbots run in loops, making users work harder, rather than simplify the experience for them.
We’ll explore how to design dialogue flows, shape voice and tone and UX guidelines for more meaningful and helpful Human-AI conversations.
When getting answers is easy, we have to figure out how to ask the right questions. They must be detailed enough and specific enough. So let’s dive into the art of designing the input UX, modes of prompting and how to constrain the AI upfront with scoping and filtering.
Good prompts don’t have to require a lot of typing. We can use structured prompts, templates and presets, daemons and personas that change the AI's perspective, task builders that construct prompts and visual canvases for intuitive interactions.
Users are often confronted with a wall of text produced by AI — and they often get lost somewhere between sentences, especially in longer conversations. How do we make output more useful?
We'll explore the need for forced ranking, the consensus meter, color-coding, style lenses and collapsible reasoning traces to help users make sense of the AI output instantly.
Great UX is always built on accessible, inclusive experiences — and AI is certainly no exception. AI UIs have critical accessibility challenges, from the inverted navigation nightmare in chatbots to noisy, polluting streaming.
And we’ll see why conversations about AI accessibility start not with the UI, but with the data that AI has been trained on.
The most frustrating part of the AI experience happens after the AI gives its first response — well, welcome to the refinement journey!
We'll explore how to make refinement less annoying and less time-consuming — with direct interactions, contextual prompts and precision knobs to get the exact result users need, faster.
As AI takes over tactical tasks, our role as designers is evolving from sketching UIs to orchestrating flows. We’ll explore what it means to direct, monitor, and intervene with AI.
We’ll see why building guardrails, permissions and approval flows is so critical. This is our new job: keeping the AI on a leash, so it is always aligned with human values.
In times when AI can act on user’s beahalf — by sending emails, spending money and initiating actions — we need to establish robust guardrails and a safety net to avoid AI agents going off the track.
Not every AI feature can deliver on its promises, and not every problem needs an AI solution. We can map out AI capabilities against the value they deliver, to prioritize, estimate and choose AI initiatives to work on. That’s usually called AI Capability/Value Matrix.
Once we identified high-value opportunities with AI, we need to start designing AI flows. How exactly would it work though? Let’s explore how design teams create, iterate on and evolve AI features from scratch — from low-fidelity AI prototypes to testing, refinements and usability testing.
How do we measure the quality of AI experiences? How do we track AI performance over time, and its impact on UX, loyalty, retention and quality of output?
Meet the world of AI Evals — a way of tracking the performance and reliability of AI systems. Let’s dive into the specifics of how we can make it work for AI features.
AI is fragile, and often mistakes aren’t an exception, but rather a matter of time. And every time a user discovers a mistake, it’s a small betrayal of trust. Mistakes are expensive as each betrayal chips away from the carefully orchestrated relationship with the user.
You’ve made it all the way to the last session! But what’s next? Just a few closing thoughts on next steps — and how to feel comfortable and confident with the ever-evolving AI. Plus, the role of humans in that AI world, and the values that we bring to the table.
10h Video, Real-Life Examples, Checklists & Live Sessions.
With 100s of practical examples, guidelines, gotchas, checklists, takeaways & exercises — applicable to your project right away.
Add-on: Live UX training. Deep-dive sessions with 1:1 feedback & UX certification.
In a 2-weeks sprint, you’ll learn how to deal with real-life AI challenges from the ground up. You’ll be designing complex AI interfaces, receive feedback and work with Vitaly and people around the world.
Twice a year, we’ll host Live UX Training with live sessions on:
How to get started with AI UX,
Live, up-to-date sessions and certification,
Real-life case studies and examples,
UX deep dives and case studies,
Design reviews and revisions
Personal 1:1 feedback with Vitaly.
You’ll have enough time to ask all your questions and learn at your own pace. And: you can get a UX certificate as well. Check the next UX training.
I’m a huge fan of Vitaly’s work for its depth, clarity and practicality, and this course is no exception. How to Design AI Interfaces That Actually Work for People cuts through the hype and shows how AI can be fast, trustworthy and genuinely helpful.
A powerful new paradigm is here, and everybody's scrambling to figure out how to wield AI. It's still early days, but one thing's become abundantly clear: CHATBOT INTERFACES ARE WEAK SAUCE. There's so much potential to improve the UX of AI and thoughtfully integrate AI into digital products. Enter Vitaly! He's spent more time than nearly anyone else on Earth sweating the UX details of digital interfaces, and he's going to bring his thoughtful, practical, and fun advice to this brave new world!
As always, Vitaly has delivered another outstanding course. While it’s titled Design Patterns for AI Interfaces, you’ll actually gain much deeper insights into AI itself. The sheer amount of information is impressive — starting with how AI works and moving all the way through design patterns, user experience, accessibility, and beyond.
As a developer, it’s so easy to get hyper-focused on the code itself and lose sight of the bigger picture. This friendly course provides a ton of “aha” moments that will help you make better decisions for the people using your sites.
Design Patterns for AI Interfaces, brought to you by your AI Mastermind from a not-so-distant future, Vitaly Friedman.
My name is Vitaly, and over the last 15 years, I’ve been scouting the web for reliable interface design patterns. I was honored to be involved in design and testing first-hand, studying solutions that worked well or failed miserably. All these learnings and insights from all these years are now collected in this video course.
Over the years, I was wearing many hats, as designer and as developer, being involved in UX, interface design, usability testing and performance optimization.
I love our wonderful community, and so love writing articles about the web, speaking, workshopping and teaching. I’m the editor-in-chief and creative lead of Smashing Magazine. In my spare time, I’m a front-end & UX consultant in Europe and abroad, working with big and small companies and organizations, such as European Parliament, Haufe-Lexware, Axel-Springer and others.
