inflearn logo

[Offline First-come, first-served] Lee Jae-gyu (Ouroboros) Builder Meet - Harness, a contract handed to an agent

What you walk away with

  • A runtime-agnostic harness design perspective that eliminates the need to relearn from scratch every time a new AI tool is released.

  • Decision criteria for solving agent divergence, false completion reports, non-reproducibility, and permission issues through structure rather than prompts.

  • A map to understanding the next trends, such as Meta-harness, Self-Improving Agents, and Agent Memory, ahead of others.

  • How to manage open source in collaboration with AI review bots: Release gates, contributor trust, metrics beyond star counts, GTM

Starts Jul 30

Finish Jul 30

Jae Gyu Lee님과 함께해요!

Career Verified

153

Learners

8

Reviews

5.0

Rating

2

Courses

Hello. I am Jaegyu Lee, and I am currently building Ouroboros.

How we can validate Agents,

I am researching how agents can remember effectively.

I am cultivating open source so that everyone can use AI technology effectively.

 


 

Lecture History

  • Seoul National University, College of Transdisciplinary Studies, Silicon Valley Design Thinking Lecture 2H (2026.01)

     

  • Seoul National University of Science and Technology ITM Department OLAP and AI Data Analysis Lecture (2026.04)

     

  • Seoul National University, School of Transdisciplinary Innovations, Lecture on Design Thinking and Socratic Reasoning in the Silicon Valley AI Era 1H(2026.04)

  • Seoul National University, College of Transdisciplinary Studies, AI Education Lecture 1H(2026.05)

  • KT AI Capacity Building Lecture, Harness Engineering: How to control stochastic AI as a deterministic system 4H(2026.05)

  • Seoul National University X San Francisco AI Native Lecture 1H(2026.06)

  • Hanwha AX Lecture 3H(2026.07)

  • KT AI Capacity Building Lecture, AI tokenomics: How to use AI tokens efficiently, 4H(2026.07)

 

External Activities

  • 1st place at Ralphton #1, Judge and Speaker at Ralphton #2

  • GDG Build with AI Incheon, Major Knowledge Required for Harness Engineering Session 1H

  • Hermes Meetup, RLM-FORGE Session 1H

  • Arize AI Meetup, Agent OS Session

  • GDG Korea-Japan Hackathon Judge

  • OBA Hackathon Award Winner

  • ICML meetup Agent Client Protocol presentation

  • Various YouTube filming (https://www.youtube.com/@Q00_Dev)

 

More

7월

30일

챌린지 시작일

2026년 7월 30일 오전 10:00

챌린지 종료일

2026년 7월 30일 오후 12:30

챌린지 커리큘럼

All

1 lectures

Recommended for
these people

Who is this course right for?

  • Developers who already use AI coding tools like Claude Code in their work or side projects, but are tired of constantly cleaning up the messes these agents leave behind.

  • A developer who switches between various CLIs and has experienced their settings and know-how being reset every time they learn a new tool.

  • Tech leads and engineering managers who need to introduce AI coding to their teams and are more concerned about "what standards to operate by" than "which tool to use"

  • Builders who want to create their own tools or products on top of agents, and developers who dream of running open-source projects.

Reviews

All

7 reviews

5.0

7 reviews

  • dolilu님의 프로필 이미지
    dolilu

    Reviews 12

    ∙

    Average Rating 4.8

    5

    100% enrolled

    I was able to gain a lot of insights into Harness and Ouroboros. It would be even better if there were longer lectures or sessions.

    • rjanan21951님의 프로필 이미지
      rjanan21951

      Reviews 1

      ∙

      Average Rating 5.0

      5

      0% enrolled

      It was a truly great time to gain insights into the harness. Even as models evolve, it seems very important to know how to use the harness properly.

      • jysohn08252864님의 프로필 이미지
        jysohn08252864

        Reviews 1

        ∙

        Average Rating 5.0

        Edited

        5

        0% enrolled

        I'm leaving with some great insights about harnesses!

        • gkwlaks님의 프로필 이미지
          gkwlaks

          Reviews 2

          ∙

          Average Rating 5.0

          5

          100% enrolled

          It was a great time gaining insights into the harness!

          • 2nan님의 프로필 이미지
            2nan

            Reviews 1

            ∙

            Average Rating 5.0

            5

            100% enrolled

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

            $23.10