inflearn logo
Challenge
Ended

[Free Live] Harness Engineering: Building an AI Development System That Runs on Its Own

Building an MVP with AI is no longer difficult. However, most projects stop at the next stage. 👉 You’ve built the features 👉 But development doesn't continue Why does this happen? The problem isn't the code; 👉 It's because there is no structure that allows AI to work continuously. --- In this challenge, based on an existing project, 👉 We will personally build a structure 👉 That allows AI to continue development indefinitely. --- Rather than simply using AI tools, we will: * Create a docs structure * Define the SSOT (Single Source of Truth) * Execute development on a per-ticket basis * Connect everything to QA and iterative cycles 👉 To complete a single "AI Development System." --- Through this process, 👉 A structure where development continues 👉 Even without a human manually coding In other words, 👉 We will make a development system 👉 That operates AI like a team actually work. --- This challenge is for: 👉 Those who want to build a structure they can apply to their projects immediately 👉 Those who started development with AI but found it difficult to sustain 👉 Those who want to move to the next level after "Vibe Coding" This is a 1-week hands-on challenge. --- Beyond just learning, we provide an experience where you: 👉 Execute 👉 Make it work 👉 And take it away as your own.

cursor
ChatGPT
AI Agent
Vibe Coding
Free Live

1개 수업 학습

2회 미션 수행

라이브 1 회

knodark74님과 함께해요!

Career Verified

540

Learners

19

Reviews

4

Answers

4.2

Rating

2

Courses

As an engineer with over 25 years of development experience,
I have built a "structure where AI can continuously carry out development"
through a repeated process of building and dismantling MVPs directly with AI.

I am directly designing and operating an "AI development system that runs on its own," covering everything from
planning → development → QA → deployment
based on ChatGPT, Cursor, and Python.

It is not just simple vibe coding;
I am creating a structure where AI Agents can continuously perform development,
and I am validating this by applying it to actual services.

More

🎉 Notice of Recruitment Capacity Increase

Due to overwhelming interest, the application closed early, so we have increased the recruitment capacity to 500 people. Thank you! 😊 😊


I built a service with AI, but..
why am I still doing everything myself??


I thought it was automated, but I'm still doing it myself?!

The problem is not the code.

You need to change the structure of how AI works.


Why is it not growing?

3 reasons why services created with AI
come to a halt 🚨

A structural issue, not a code quality issue

Even if AI writes code well, without an overall structure, the service will quickly hit its limits.

Disconnection between QA, Operations, and Deployment

Because each stage is not connected, if one thing changes, the entire system must be overhauled.

Constant human intervention is required

It is called automation, but in the end, it only works if a human is involved in the middle.

Core Concept

What is Harness Engineering?

Harness Engineering is not about writing code directly, but rather
a method of designing an environment where AI can work reliably.

"Now is the time to change the way we develop."

The role of the developer is changing. Now, a developer is not someone who creates code, but someone who builds the structure in which AI works. In this live session, I intend to show you the possibilities and know-how of how to build that structure based on real-world cases.



Live Highlights

What we cover in this live session

01

Agent-first development

Transition to a development workflow driven by AI agents.

02

SSOT-based Structure

Unify the structure with a Markdown-based Single Source of Truth (SSOT).

03

Ticket-driven development workflow

Generate tickets with GPT and implement them automatically with Cursor.

04

Automated QA & Done

It automatically handles everything from generating analysis reports to verifying completion criteria.

05

Sustainable Development Loop

Once set up, it establishes a repeatable automated development cycle.


Live Demonstration

We will show you directly in the live demo.

STEP 01

📝 Ticket Creation

  • Markdown-based ticket creation

  • Structuring with GPT

STEP 02

⚙️ Development Automation

  • Cursor-based implementation

  • Code generation and modification

STEP 03

✅ QA & Completion

  • Generate analysis report

  • Verification of Done Criteria


The automation flow we will experience

In the service development process, humans do only two things — defining what to build, and verifying the results.
Everything else, except for ticket creation and QA, is performed by AI.


Tech Stack to Apply

  • GPT (Planning / Ticket Creation)

  • Cursor (Development / Modification)

  • Markdown-based SSOT

  • MongoDB / API Structure

  • Oracle VM Deployment

The core is the structure, not the technology.
The principles covered here can be applied to any stack.

Target Audience

Recommended for the following people:

✅ Those who have built services with AI but find operations difficult

✅ Those who want to handle everything from development and QA to deployment on their own

✅ Those who want to properly utilize Cursor and GPT

✅ Developers curious about development methods in the AI era


This live session,

Once you see it,
you won't be able to go back to the old way.


Amidst the changing development methods in the AI era

I will show you the actual development system where AI works,
from start to finish.

4월

9일

챌린지 시작일

2026년 4월 9일 AM 10:00

챌린지 종료일

2026년 4월 9일 PM 12:00

챌린지 커리큘럼

All

4 lectures ∙ (2hr 34min)

Course Materials:

Lecture resources
Live

Reviews

All

9 reviews

4.7

9 reviews

  • aseltula5967706님의 프로필 이미지
    aseltula5967706

    Reviews 11

    Average Rating 5.0

    5

    100% enrolled

    The materials were extensive, and I liked that you provided detailed answers to my questions.

    • dooee님의 프로필 이미지
      dooee

      Reviews 1

      Average Rating 5.0

      5

      100% enrolled

      It was a wonderful lecture. I'll be praying for it to be officially organized(?) as a separate course.

      • lyw311362776님의 프로필 이미지
        lyw311362776

        Reviews 1

        Average Rating 5.0

        5

        0% enrolled

        This was a truly useful lecture. The quality is amazing!!

        • korean12434943님의 프로필 이미지
          korean12434943

          Reviews 1

          Average Rating 5.0

          5

          0% enrolled

          The instructor was very kind in answering questions from many students, and the quality of the lecture was high and excellent!

          • thdalsrua127526님의 프로필 이미지
            thdalsrua127526

            Reviews 3

            Average Rating 5.0

            5

            100% enrolled

            It was very helpful that you focused on hands-on practice by showing us how you do it yourself, and sharing the docs file was a huge help. Thank you!

            취소 및 환불 규정
            챌린지는 지식공유자가 설정한 수업 최소 정원이 충족되지 않을 경우, 폐강 안내가 고지되며 결제 내역이 자동취소됩니다.

            knodark74's other courses

            Check out other courses by the instructor!

            Free