All classes Claude from Zero to Expert Syllabus · Episode 96 of 100

Claude from Zero to Expert Class #96: Choosing a Realistic Final Project

0 views

Choose a realistic final project: the traits of a project that can be finished and actually used, sample ideas for businesses, offices, teachers, and freelancers, how to score ideas, defining must-have features, then writing a project sheet and CLAUDE.md with Claude. Raka chooses a mini cashier app for Sunset Coffee. Episode 96 of 100, Session 20: Final Project.

Software: Claude
Level: Advanced
Duration: ± 8 min
Series total: 15 h 1 min · 100 videos
Published: October 28, 2026

Episode 96 of 100 · view syllabus

Download PDF (5 PDF pages)

0 downloads

Unlock All Premium Classes ▶ Watch the video lesson0 plays

▶ Video Lesson

Watch every step above performed live, with English narration.

0 plays

Class book

Page 1 / 5

Claude from Zero to Expert Class #96: Choosing a Realistic Final Project: Cover: Claude from Zero to Expert Class #96: Choosing a Realistic Final Project

Claude from Zero to Expert Class #96: Choosing a Realistic Final Project: Before you start, step list, and Step 1: Traits of a good final project

Claude from Zero to Expert Class #96: Choosing a Realistic Final Project: Steps 2–3: Sample ideas from our world

Claude from Zero to Expert Class #96: Choosing a Realistic Final Project: Steps 4–5: Writing a project sheet

Claude from Zero to Expert Class #96: Choosing a Realistic Final Project: Done: summary and further practice

The end

Cover: Claude from Zero to Expert Class #96: Choosing a Realistic Final Project

Click a page or use ◀ ▶ to turn · swipe on mobile · to enlarge

Want to go deeper?

Join one of our courses or ask us directly on WhatsApp.

Download PDF· 0 downloads View the full 100-episode syllabus View Claude AI Courses Ask on WhatsApp

Recommended classes for you

Picked from the software, topic, and level of “Claude from Zero to Expert Class #96: Choosing a Realistic Final Project”.

50 episodesFree preview

Agentic AI · Beginner–Intermediate

Agentic AI Without Coding

50 video lessons with PDF guides, 14 h 29 min in total. Starts with: Chatbots, Automation, and AI Agents: What's the Difference?.

14 h 29 min 66 0

View class

18 episodesFree preview

Python · Beginner–Advanced

Machine Learning Python

18 video lessons with PDF guides, 3 h 41 min in total. Starts with: Introduction to Machine Learning & First Model.

3 h 41 min 61 0

View class

6 episodesFree preview

Python · Intermediate–Advanced

Brain Tumor Segmentation with U-Net

6 video lessons with PDF guides, 1 h 10 min in total. Starts with: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Splitting.

1 h 10 min 28 0

View class

Deep Learning CNN-LSTM Pneumonia Detection

6 episodesFree preview

Python · Intermediate–Advanced

Deep Learning CNN-LSTM Pneumonia Detection

6 video lessons with PDF guides, 57 min in total. Starts with: Chest X-ray Dataset: Loading, Exploration & Data Splitting.

57 min 22 0

View class

Need more than a video?

Other ways we can help with Claude.

Learn with a mentor

Scheduled AI & Machine Learning courses with live guidance and a certificate.

Course

Midjourney Course: Create Stunning AI Images from Text

★ 4.9

Course

DALL-E Course: Creating Professional AI Images from Text

★ 4.9

Course

Stable Diffusion Course: AI Art from Beginner to Professional

★ 4.9

See all in AI & Machine Learning →

Have it built for you

No time to learn Claude? Our team can do the project for you.

Service

AI Image Creation with Stable Diffusion — Unlimited Custom Visual Production

★ 4.9

Service

AI Image & Video Creation with Leonardo AI — All-in-One Visuals for Business

★ 4.9

Service

AI Image Creation with Midjourney — Cinematic AI Art for Business

★ 4.9

See all in AI & Machine Learning →

Get notified about new Claude classes

At most one email a week. Unfollow anytime with one click.

Transcript

The full narration of this video. The current sentence is highlighted while the video plays; click any sentence to jump there.

Show transcript

Introduction

Hi, and welcome to the Final Project session. Over nineteen sessions we've learned everything from our first chat to bots and agents. Now it's time to bring it all together into one project that people actually use. In this episode we don't write code yet. We choose a realistic idea, assess whether it's feasible, define the must-have features, then write a plan for Claude Code.

Step 1: Traits of a good final project

A good final project has five traits. First, it solves a real problem that happens every day or every week, not one that comes up once a year. Second, it has a clear user. The person who uses the result will give feedback that keeps making the project better. Third, it's small enough to finish in one to two weeks. A giant project that never gets finished brings no benefit at all. Fourth, it can be tested with sample data, so we know for sure whether the result is correct. Fifth, it's safe. If the project touches money or personal data, the safeguards are designed from the start, as we learned in the API session. Measure your idea against these five traits. If two or more aren't met yet, shrink the idea first. For example, not a complete school app, just a question bank for one subject. There are also three traps. Trap one: too big. You want to build an app right away that's as complete as one from a big company. Trap two: busy polishing the looks, while the main function doesn't work yet. Trap three: no users. The project is finished, and then nobody opens it. This is the saddest trap, and the easiest to prevent by asking your future users first. None of these traps is about technical skill. With Claude Code, building features feels easy, so the temptation to add more only grows. The discipline to choose what's essential is the key skill in this session.

Step 2: Sample ideas from our world

Let's look at sample ideas from our characters. Laras still tallies orders on paper, and the sales recap every night takes a long time. The idea: a mini cashier app built with Claude Code. Dewi at Maju Bersama Co. compiles the weekly stock report by hand. The idea: an automatic stock report script, like in the automation session. Mr. Hendra writes practice questions every week. The idea: a question bank with an answer key, using a Project and an Artifact. Nadia writes captions and schedules for five clients. The idea: an automatic content calendar with n8n and the Claude API. The pattern is always the same: one recurring problem, one user, one tool from this class. Now write down the equivalent for your own work. To make sure the idea is clear, use the one-sentence project formula. Part one, the user: for Laras and Dimas at the Sunset Coffee shop. Part two, the problem: who struggle with making change and a thirty-minute nightly recap. Part three, the tool: I'm building a mini cashier on the shop's laptop with Claude Code. Part four, the measure of success: so the nightly recap takes five minutes at most, with no miscounts. If you can't fill in all four parts for your idea yet, the idea isn't ready to build. Usually what's missing is the user or the measure of success, and you can get both by asking your future users directly.

Step 3: Scoring ideas

Raka has three ideas for Sunset Coffee and scores them on four measures, from one to five. Idea one: a full delivery app with online payments. The impact is big, but it's almost impossible to finish in two weeks, and the risk is high because it involves money. Idea two: a WhatsApp agent that takes orders on its own. Interesting, but there's still a risk of recording orders wrongly without supervision. Idea three: a mini cashier on the shop's laptop with a daily recap. The impact is still big, it can be finished, the data already exists in the menu list, and the risk is low because it's used inside the shop. Idea three wins. Not the coolest one, but the most likely to be finished and used. Ideas one and two aren't thrown away; they become follow-up plans once the cashier proves useful. Before deciding, Raka asks the future user directly. He asks what the biggest hassle in the shop is right now. Laras answers: making change when it's busy, and the nightly recap that keeps Dimas busy for thirty minutes every night. Raka then asks the most important question: when would this cashier count as a success? Laras's answer becomes a clear measure of success: the nightly recap done in five minutes, with no more miscounts. This is the measure we'll test later. This short chat also makes Laras feel involved. Users who help set the measure of success are usually more patient when testing and more diligent about using the result.

Step 4: Writing a project sheet

Now we write the project sheet. Raka opens Claude and explains the problem and Laras's measure of success. Claude lays out the project sheet in the Artifact panel. Look at the table: must-have features on the left, later features on the right. There are only four must-haves: picking menu items with an automatic total, calculating change, recording the payment method, and a daily recap. Raka adds one important limit: QRIS, Indonesia's QR code payment, is only recorded as a method, not connected to the bank. Connecting to a payment system is a big job with high risk. Claude updates the sheet to version two. The schedule is clear too: design for one day, build for three days, test for three days, and document for one day. Later features aren't thrown away. Delivery, ingredient stock, monthly reports, and a WhatsApp agent wait until the cashier proves it's used every day. A small project that's finished is always worth more than a big project that's half done.

Step 5: From project sheet to CLAUDE.md

Raka creates a new folder called sunset coffee cashier, saves the project sheet there, then asks Claude Code to write CLAUDE.md. Notice the limit: don't write any app code yet. Claude Code reads the project sheet first. Then it writes CLAUDE.md, about thirty-two lines long. The content is concise: the goal of a cashier for one laptop that works without internet, four must-have features, a ban on connecting to the bank or storing customers' personal data, and a rule that prices come only from the menu file. Claude Code reads this file every time it works, so it won't add features outside the plan. Raka also asks what he needs to prepare before building starts. Asking like this prevents surprises along the way. Claude Code names three things: a menu file with four products and Laras's official prices, a check of the browser on the shop laptop, and ten sample transactions for testing, without customer names. Raka prepares all three before the next episode. Here's the plan for our last session. Today we chose a project and designed it. Next, we build the Sunset Coffee mini cashier with Claude Code. Then we test and fix it until Dimas is ready to use it at the shop. After that we write the documentation and prepare a project presentation. The screens in this video are simulations, so menu names may differ slightly from yours.

Done!

We've learned five traits of a realistic project, scored ideas, asked the user, written a project sheet, and turned it into CLAUDE.md. Now it's your turn to choose your own project. In the next episode, Raka starts building the Sunset Coffee mini cashier app with Claude Code, from designing the interface to a version you can try out. See you there!

Practice exercises

Try these on your own after watching to lock in what you learned.

Score three of your project ideas with a table for impact, size, data, and risk
Create a project sheet for the chosen idea together with Claude
Ask Claude Code to write CLAUDE.md from that project sheet

Ask the class material

Type a question or keyword; we find the part of this episode that explains it and jump the video there.

My notes

Notes are saved with the current video time and are only visible to you.

Questions & answers

Stuck on a step? Ask here; the instructor and other learners can answer.

No questions yet.

Your instructor

Cerdas Inovasi Digital Team

Software developers, BIM & 3D engineers, and AI practitioners

53 video classes1,296 episodes

We build software, BIM models, games and AI automations for clients every day, and turn that hands-on work into step-by-step classes: every step is shown on screen and comes with a PDF guide you can keep.

See our clients →

Memuat

Claude from Zero to Expert Class #96: Choosing a Realistic Final Project

0 views

Software

Claude

Level

Advanced

Duration

± 8 min

Series total

15 h 1 min · 100 videos

Published

October 28, 2026

Episode 96 of 100 · view syllabus

Unlock All Premium Classes

▶ Watch the video lesson0 plays

Transcript

The full narration of this video. The current sentence is highlighted while the video plays; click any sentence to jump there.

Show transcript

Introduction

Step 1: Traits of a good final project

Step 2: Sample ideas from our world

Step 3: Scoring ideas

Step 4: Writing a project sheet

Step 5: From project sheet to CLAUDE.md

Done!

Your instructor

Cerdas Inovasi Digital Team

Software developers, BIM & 3D engineers, and AI practitioners

53 video classes1,296 episodes

See our clients →

▶ Video Lesson

Recommended classes for you

Need more than a video?

Transcript

0:03Introduction

0:27Step 1: Traits of a good final project

1:59Step 2: Sample ideas from our world

3:33Step 3: Scoring ideas

5:10Step 4: Writing a project sheet

6:15Step 5: From project sheet to CLAUDE.md

7:45Done!

Practice exercises

Ask the class material

My notes

Questions & answers

Your instructor

▶ Video Lesson

Recommended classes for you

Need more than a video?

Transcript

0:03Introduction

0:27Step 1: Traits of a good final project

1:59Step 2: Sample ideas from our world

3:33Step 3: Scoring ideas

5:10Step 4: Writing a project sheet

6:15Step 5: From project sheet to CLAUDE.md

7:45Done!

Practice exercises

Ask the class material

My notes

Questions & answers

Your instructor

Introduction

Step 1: Traits of a good final project

Step 2: Sample ideas from our world

Step 3: Scoring ideas

Step 4: Writing a project sheet

Step 5: From project sheet to CLAUDE.md

Done!

Introduction

Step 1: Traits of a good final project

Step 2: Sample ideas from our world

Step 3: Scoring ideas

Step 4: Writing a project sheet

Step 5: From project sheet to CLAUDE.md

Done!