1.7.3. Final Assessment, Part 1 — Transcript

Detailed walkthrough of the final capstone project for AI entrepreneurship, focusing on inquiry-based learning and startup strategy.

Key Takeaways

The final project is designed to simulate launching a real startup by examining it from multiple critical angles.
Inquiry and asking original, context-specific questions are central to innovation and startup success.
The Inspire company serves as a blueprint for structuring ventures, not as a business model to copy.
Students must engage deeply with each phase, creating their own questions and strategies.
Understanding the reasoning behind each phase leads to higher quality and more impactful work.

Summary

The video introduces the final capstone project in the AI-SANA program, emphasizing its importance beyond just earning a certificate.
The project requires students to stress test a real business idea from 12 different critical startup dimensions.
The core philosophy is Mobile Inquiry Based Learning Environment (Smile), which values asking innovative questions over memorization.
A fictitious company called Inspire is used as a sample model to demonstrate the structure and logic students should replicate.
Students are encouraged to form diverse teams and write high-level questions that meet Bloom's taxonomy level four or higher.
The video guides navigation of the Smile platform, especially the group section where the Inspire model is hosted.
The 12 phases of the project cover essential startup aspects such as vision, market landscape, user segmentation, and motivation design.
Each phase requires students to generate 20 original questions tailored to their own venture idea, not copying samples.
The video stresses thinking like a founder or chief strategy officer, focusing on deep inquiry and systemic startup thinking.
The project aims to prepare students to launch a real venture by applying structured, critical, and innovative thinking.

Full Transcript — Download SRT & Markdown

Speaker A

Hi everyone, welcome back and welcome to what is arguably the most important video in this entire program. If you made it here, that means you've already absorbed a significant body of knowledge about AI entrepreneurship and the future of work. And now it's time to pull all

Speaker A

of that together. In this video, I'm going to walk you through the final capstone project in complete detail. I want you to understand not just what to do, but why we designed it this way.

Speaker A

Because when you understand the reasoning, you will do extraordinary work. This project, which you will find in the Smile platform as your final module, is a mandatory requirement to complete the program and earn your certificate. But I want you to hear this

Speaker A

clearly. The certificate is not the point. The certificate is a byproduct. The real point is that by the time you finish this project, you will have done the intellectual work of actually launching a company. Not pretending, not theorizing, actually stress testing a

Speaker A

real business idea from 12 different angles. We want you to stop thinking, uh, like a student sitting in a classroom and start thinking like a founder, like a chief strategy officer, like someone who has skin in the game. Because in the real

Speaker A

world, the quality of your output is always determined by the quality of your input.

Speaker A

And in the world of startups, that input begins with one word: inquiry. This is the core philosophy of the Smile platform: Stand for Mobile Inquiry Based Learning Environment. The belief is that the most powerful thing a human being

Speaker A

can do is ask a great question. Not memorize an answer. Not repeat what a professor said. Ask a question that nobody has asked before about your specific problem in your specific context for your specific user. That's where innovation is born. By the end of

Speaker A

this session, you will understand exactly how to navigate the Smile platform for this project. How to study our sample Inspire.

Speaker A

This is a fictitious company model. Um, and I'll talk about that in detail. This will give you an idea how to form a high-performing diverse team. How to write questions that reach Bloom's taxonomy level four or higher,

Speaker A

which, um, is what our AI evaluator is looking for, and how to think through your startup as a complete living system rather than just a cool idea. Let's get to it. Before we write a single question, I want you to do one

Speaker A

thing first. I want to, uh, I want you to study the roadmap. Go to Smile platform like you're looking at here and by now you've logged in many times, right, to complete many activities. So you know

Speaker A

the interface well and navigate to the group section. So this is where we want to be, but we'll start from groups. Okay.

Speaker A

Go to groups and then, uh, you will type in buyer. Probably you will have a link to this group anyway in the syllabus. So, uh, it's so easy to find, but if you don't have that link, you can just go to

Speaker A

groups and then search for Inspire and here it is. That's it. And then you click on the name and you will go into that group and you are back here, right?

Speaker A

Okay. So, uh, Inspire is a fictitious company we built specifically for this program. It's an AI-driven self-coaching platform designed for students. It uses machine learning to analyze each learner's unique patterns. How they study, when they disengage, what emotional states correlate with their

Speaker A

best performance, and it builds what we call a growth portfolio for each user. It's an ad tech company at its core, but its function here is to serve as a sample model for your own venture. Now, I want you to be very clear. You are not

Speaker A

here to build an ad tech company. You are not here to copy Inspire. You might want to build a healthcare company or fintech solution or something in agriculture technology or green energy or logistics. The industry does not matter.

Speaker A

What matters is the structure of how Inspire is organized. Every phase, every question, every layer of, uh, thinking that Inspire models, that is what you need to replicate for your own venture.

Speaker A

Think of Inspire the way an architect thinks about a blueprint. A blueprint for a hospital and a blueprint for a school look very different on the surface, but they follow the same engineering logic. Load-bearing walls, electrical systems, fire exits. Inspire is your engineering

Speaker A

logic. Okay. Your job is to apply that logic to your own building. Spend some time in the Inspire group. Read the questions.

Speaker A

Notice how they are structured. Notice that they are not simple. Notice that they don't just ask what. They ask how and why and in what context, under what conditions.

Speaker A

This is the language of a sophisticated thinker. That is the language we want from you. Now let's look at the architecture of the project itself. In your group on Smile, you'll find 12 phases. As you can see right here, 12

Speaker A

distinct activities, each representing a critical dimension of building a real venture. I want to walk you through each one briefly so that you understand what territory you're covering. Okay. Uh, each phase exists because real startups fail when they neglect it. So first phase

Speaker A

phase one: vision, philosophy, problem reframing. This is where you answer the existential questions.

Speaker A

Why does this company exist? Not because there's a market opportunity that's too thin. What does it matter? What is the change you want to see in the world? And crucially, how do you reframe the problem so that your solution feels

Speaker A

inevitable rather than optional? The best founders don't just solve problems, they reframe problems so that the solution becomes obvious.

Speaker A

Here are some sample questions. Do not copy these questions. I want you to come up with your own question for this particular phase. Phase one, vision, philosophy, problem reframing.

Speaker A

You can take a look at these questions and see what, uh, questions are there for this particular phase. Uh, and I want you to do the same. Have your own questions generated for this phase. Okay? Uh, exactly tw, uh, 20 of them. Got it. The

Speaker A

second phase, market landscape and competitive positioning. Who else is out there and more importantly, what do they think the problem is? Because your competitive advantage often lives in the gap between what incumbents believe and what users actually experience. This phase forces

Speaker A

you to study the battlefield before you enter it. So you can take a look at these sample questions. Again, do not copy these questions as they are. Come up with your own set of questions, 20 questions with your own starter venture

Speaker A

idea. Got it? So phase three: target users and segmentation strategy. Who exactly are you helping? Who are you really helping? Everyone is not an answer. Young people is not an answer.

Speaker A

You need to go deep. What does your user's day look like at 7 in the morning?

Speaker A

What problem are they trying to solve before breakfast? What, uh, great segmentation is not demographic, it's psychographic, it's behavioral, it's situational. So again, great segmentation is not demographic. It's psychographic. It's behavioral. It's situational. And these questions cover what sort of questions

Speaker A

you of a company, a startup, a venture might ask. Again, do not copy this question. These are just the sample questions that you might ask, but have your own questions posted, 20 of them for phase three. Okay? And then phase

Speaker A

four: motivation design, engagement psychology. This is a phase that most startups skip entirely and it's why most products get downloaded and never opened again. How do you design for sustained motivation?

Speaker A

What psychological principles drive continued engagement? This is where behavioral economics meets product design. You need to ask hard questions about habit formation, intrinsic versus extrinsic reward, and the emotional journey of your user. Again, here are some 20 questions that are

Speaker A

sample for you to look at and reference. But do not copy those questions. You come up with your own 20 questions with your team members.

Speaker A

Okay. Phase five: AI infrastructure, personalization, self-coaching engine. We are

Speaker A

the experience how does it learn from user behavior and get smarter over time this phase requires you to think like a technical architect. Even if you are not an engineer, you don't need to write code, but you need to understand the

Speaker A

logic. If you do not know how to write a code, get someone who can write a code.

Speaker A

And by this time, you know how to work with AI resources and tools to write some codes. Probably uh you can use some vibe coding tools, right? and use a replet or lovable or uh cloud code, you can have AI write the code for you.

Speaker A

Okay? And by by this time you probably have run many Python codes already to finish this program. So you probably feel comfortable doing that. Okay. But here are the questions you may want to ask in this phase, phase five. and then

Speaker A

um post your 20 questions for phase five. Phase six, product experience, UX, UI, prototyping.

Speaker A

What does your product actually feel like to use? This is where vision meets reality, great ideas fail because the interface is confusing or the on boarding is too long or the design doesn't match the cultural expectations of the user. In Kazakhstan, for example,

Speaker A

your users may interact with your product primarily on a mobile phone in conditions very different from a Silicon Valley office or users sitting in Zanzibar, Tanzania or South Korea or China, right? Different users use different uh devices uh as a primary

Speaker A

device to interact. So come up with your own 20 questions and this regards to design for your user in uh in your users's context in the country in the region they they are located. Okay. So it's got to be localized and

Speaker A

culturalized. Phase seven business model revenue architecture and pricing. This is very important as well. Every phase is very important. How does your company make money? Right? You got to make money, right? It's a it's a for-profit venture. This sounds basic,

Speaker A

but it's surprisingly complex. Are you subscriptionbased? Transaction based. Do you have a preminium tier? Premium, premium. You heard that already. What's your unit economics? meaning how much does it cost you to acquire one user and how much revenue does that user generate

Speaker A

over their lifetime. These numbers tell the truth about whether a business is viable or not. Here are some questions, sample questions you can take a look at.

Speaker A

Do not copy these questions. Have your own questions posted for this phase seven as well. Now phase eight, funding, investor strategy, financial planning. Where does the capital come from to build this? Bootstrapping, angel investors, venture capital, uh, government grants. Kazakhstan has a

Speaker A

growing ecosystem of innovation funds and international development partners. Are you aware of them? How do you tell your financial story to someone who is about to write your check? This phase teaches you to think like an investor about your own company. And as you can

Speaker A

see, some of these questions probably relevant for you. Okay? Or you probably come up with a different set of questions regarding this particular phase of funding. Okay? Uh phase then let's go to the next phase. Phase nine.

Speaker A

What's in phase nine? Team, talent, organization, design. Very important. A startup is not a product. A startup is a team that builds products.

Speaker A

How do you structure your organization? What roles do you hire for? First, how do you build a culture that attracts people who want to do meaningful work?

Speaker A

And critically, how do you lead a diverse team through the inevitable moments of crisis and disengagement?

Speaker A

These things happen in the real world. So, you probably have your own set of questions you want to ask. So, post them 20 of them. Phase 10, go to market and branding launch strategy. You can build the most brilliant product in the world

Speaker A

and it will die in silence if nobody knows it exists. How do you enter the market? What does your brand communicate? What is your first 90-day launch plan? Who are you? Your first 100 users and how do you find them? This

Speaker A

phase is about the art of making noise in a crowded world. You want to get noticed, right? So you probably have your own set of 20 questions and then do not copy these questions as they are but I hope that you get some ideas and get

Speaker A

inspired to come up with your own questions. Phase 11, metrics, validation, continuous innovation. Once you launch, how do you know it's if it's working? What metrics matter? How do you distinguish between it between a real signal and noise? And how do you build a

Speaker A

culture of continuous experimentation where failure is uh data and not disaster? This phase is about building the nervous system of your company.

Speaker A

Okay? And then here are some of the questions. All right? Um some of them are basic, some of them are very very high u level questions. But these questions are important and you do not want to miss these questions when you

Speaker A

are uh forming your startup. Phase 12, the last stage, scalability, ethics, long-term impact. Finally, what happens when it works? How do you scale from 1,000 users to 1 million without breaking what made you special? And perhaps most importantly,

Speaker A

what is your ethical obligation as a company that uses AI and data to influence human behavior? What is the long-term impact you want to have on society, not just your profit and loss statement? This phase asks you to think

Speaker A

like a statesman not just a business person. Why do we require all 12 uh phases?

Speaker A

Um because in the real world startups fail not because the founders lacked passion or intelligence. They failed because they forgot to ask a question in one of these phases. They built a beautiful product and forgot to figure out who would pay for it. They raised

Speaker A

money and forgot to think about the ethics of their data practices. By generating questions across all 12 dimensions, you are doing what we call a premorton. If you are identifying where you might fail before you even start, that is the mark of a sophisticated

Speaker A

founder. Okay. So now let's talk about the actual task. For each of these 12 phases, your team must generate 20 high quality questions. That is 240 questions total. And I want to spend time I want you to spend time here because the

Speaker A

quality of those question is everything. There is a spectrum of question quality. At the low end, you have simple recall or observation questions. At the high end, you have what Bloom's Texonomy calls level four analysis. Our AI evaluator of smile platform is

Speaker A

specifically trained to detect where your question falls on that spectrum and we'll score it on a scale of 1 to 10. So if you come up u with a question like this just click on a question like this

Speaker A

and you will have your own question. It tells you what Bloom's tonomy level is.

Speaker A

Level six. This is a level six question. The overall score is 8.2 two and confidence level is 90%.

Speaker A

Okay. Uh so if you're in phase one vision and problem reframing and you write uh the question like can drones fly over uh wheat fields you will get a very low score because that's that's not a high quality question and you uh

Speaker A

maybe two or three. Why? Because that question requires no analysis. Can drones fly over wheat fields? What kind of question is that? Oh, what is a wheat field? You know, that question will probably get score one out of 10. It

Speaker A

doesn't surface a hidden assumption. It doesn't prepare you for a decision you will need to make. Now, compare that to this question. How do we reframe the role of a traditional Kazak farmer from a manual labor to a deltadriven fleet

Speaker A

commander? through the use of a swarm intelligence and what narrative shift does that require in our marketing, our training programs and our government relations strategy. A question like that probably get nine. Why? Because it requires you to think across multiple dimensions

Speaker A

simultaneously. Culture technology communication strategy, policy. It doesn't just uh identify a feature. It identifies a transformation.

Speaker A

Uh or or consider this one for the same drone startup this time in phase 12.

Speaker A

Scalability, ethics, long-term impact. Um uh what else? Scalability and long-term effect. And then uh uh uh other uh dimensions. So for example, if you ask questions like as our um uh drones fleet begins to collect years of granular crop

Speaker A

yield data across Kazakhstan's regions, what ethical framework governs data ownership? Does that data belong to the farmer, to us, to the government? That's a very complex question, isn't it? And how do we ensure that our AI recommendations don't inevitably

Speaker A

uh create uh a new form of agricultural dependency that disadvantages smallcale farmers? Wow, what a question. That is a five-star question. It anticipates a problem that most teams won't think about uh until it's too late. And that is the point. A practical guide. If your

Speaker A

questions are starting with what is or does our does the company or can we they're probably too low level. Push yourself to ask how does and under what conditions and what is the systematic excuse me what is the system uh

Speaker A

systematic consequence of and how do we balance X against Y when those sentence um starters are signals that you are operating at the right altitude. Got it?

Speaker A

So let me talk about the team uh formation and diversity as well. Now you you are supposed to put 240 uh questions uh over 12 uh phases.

Speaker A

I recommend a team approximately seven people. Uh it could be five, six or eight, 10 but uh somewhere uh seven would be nice. But the composition of that team matters enormously. And I want to give you my most important piece of

Speaker A

advice for this entire project. Diversity is your greatest strategic asset. If you form a team of seven software engineers, you'll produce 240 engineering questions and they will be good questions, but they'll miss an entire universe of considerations.

Speaker A

You won't ask how the farmer in the Akamola region will interact with your interface when the temperature is minus 20 and they're wearing globes. You won't ask what the Ministry of Agricultures procurement process looks like or how long it takes. You won't ask what

Speaker A

happens to your brand when your drone causes an accident in a rural community that didn't consent to being part of a technology pilot. You need someone who understands the business model. You need someone who understands the technical architecture. You need someone who

Speaker A

understands uh your end user on a human cultural level. You need someone who thinks about regulation and risk. You need someone who thinks about design and communication. This diversity of thought creates what I call innovation, innovative friction. The productive

Speaker A

tension that happens when people with different mental models sits in the same room and try to agree on the right question. That friction is where your best questions will come from.

Speaker A

When you recruit your team, don't recruit friends who think the same way you do. Recruit people who will challenge your assumptions. Recruit people who have lived experiences you don't have. That uh discomfort in the brainstorm room is not a problem. It's

Speaker A

the product. Okay. So, uh the steps to do this. Let's bring this home with your technical uh marching orders. Here's exactly what you need to do in Smile platform. First go to groups, right? Click on groups and click create

Speaker A

group. Name it. Give it a name. Uh I will say Atlas. It's just a fictitious name. Atlas Momentum.

Speaker A

This is a name of a group. Okay? Whatever that may be for your startup venture. It could be a name of your startup venture. Give it a name. Give it a description.

Speaker A

Okay? And then contact information. If you want to share with the public, click on public. If you want to keep it private until you feel comfortable to share, then click on private and create group. That's how you do it.

Speaker A

Once the group is created here at last moment, that's it. And have your team members join this group. So about seven, seven would be nice. And then create activity, right? Click on create activity. Give it a name. This is like phase one, right?

Speaker A

And then that's where you put in uh the each phase, right? Phase one vision, philosophy and problem refraraming, right? That's the first one. And then you can give it a description. Um, and then just create activity. Right? So

Speaker A

once you create like an activity like here, this is where you brainstorm with your team members and then add a question, right? And then so 20 of them for each stage, each phase. So we just created one phase

Speaker A

one activity. So that's how easy it is for you to do that. Okay? And so you will create 12 of these fill them fill each phase with 20 questions your own team's question. Do not copy paste questions from the inspire the

Speaker A

fictitious company model. Okay? And then when you put in your question and then if you uh get your question evaluated by AI here you will know whether the question qualifies.

Speaker A

Like I said the broomstamy level has to be four or higher and then the AI the evaluation score should be five or higher. Okay. 20 original high quality questions for each stage. So you will be creating 12 phases here. Copy the names

Speaker A

from the company inspire. Okay. And then um create them and that's will fulfill your capstone project requirement for module 1.7. So when your team has 12 solid well-rated questions for all 12 phases, you have completed the capstone. You will receive your uh

Speaker A

program certification. Um, but more importantly, you will have done something that the vast majority of aspiring entrepreneurs never do. You will have thought all the way through your company before you spent a single dollar building it. You'll save a lot of

Speaker A

money by doing this exercise. You will have asked the questions that protect you and you will walk out of this program not as a student who learned about startups but as founder who has already begun. I cannot wait to see what

Speaker A

you build whether it's edtech or drone swarms or something none of us has imagined yet. The world needs your inquiry. World needs your questions.

Speaker A

Ask as many questions as possible. Ask more than 20 if you have more. And the more questions you have, the better prepared, the better protected your startup venture will be. Okay, best luck to you all and I'll see you um in

Speaker A

the

Topics:AI entrepreneurshipfinal assessmentcapstone projectstartup strategyinquiry-based learningSmile platformbusiness modelfounder mindsetBloom's taxonomyventure building

Frequently Asked Questions

What is the main purpose of the final capstone project in the AI-SANA program?

The main purpose is to have students apply their knowledge by stress testing a real business idea from 12 critical startup perspectives, simulating the process of launching a real company.

What is the Smile platform and how is it used in the project?

The Smile platform is the learning environment where students access the final project modules, form groups, and study the sample company Inspire to guide their own venture development.

Why does the video emphasize asking original questions rather than copying sample questions?

Because innovation arises from inquiry, students must create unique, context-specific questions that reflect their own venture’s challenges, which leads to deeper understanding and better outcomes.

Get More with the Söz AI App

Transcribe recordings, audio files, and YouTube videos — with AI summaries, speaker detection, and unlimited transcriptions.

App Store Google Play

Or transcribe another YouTube video here →