AWS deployments that can be applied directly to practice

Hello! As I previewed when announcing the publication of my first book, the 'AI Agent Evaluation' advanced course has finally opened!🎉

As mentioned in my previous message, if the book established the 'theoretical foundation and core principles' of evaluation, this lecture is a complete hands-on and application guide on how to automate and build this into a system in a practical work environment.

As you can see in the attached table of contents, this course is packed with essential content for managing agents through objective metrics that go beyond subjective testing.

📌 Key Lecture Content

• Everything about building a Golden Dataset: Efficient evaluation dataset generation techniques using the RAGAS library, LangSmith, custom agents, and Claude Code Agent Skills

• Multi-faceted Evaluation Practice: From e2e evaluation checking the accuracy of the final answer, to Component Evaluation verifying whether documents were retrieved correctly and whether tools were used in the proper sequence (Trajectory).

• Applying Advanced Evaluation Metrics and Frameworks: Anthropic's recommended agent evaluation metrics, and methods for measuring AI potential (pass@k) and consistency (pass^k)

• How to build a sustainable evaluation system

This will be the most reliable guide for those who want to go beyond simply 'implementing' an agent and instead 'prove' and optimize whether their AI "always works correctly" at a production level using data. If you have built a solid foundation through the book, now experience the 'completion of evaluation' firsthand through this follow-up practical course!

We are currently offering a 30% early bird discount to celebrate the launch, so experience the 'completion of evaluation' right now based on the solid theory you built through the book!

https://inf.run/k5fDe

Hello, I'm Kang Byeong-jin.

Hello, I recently had a great opportunity to participate in And Studio's 'Ace Report' to share my career story.

I'm facing a new beginning, but rather than this video covering the know-how of transitioning to big tech companies, it has become a space where I calmly share the thoughts I've pondered and experienced while approaching work over the years.

Looking back, I think my career has been closer to failure than success. There were experiences of starting and closing businesses that aren't recorded on LinkedIn, and moments of rejection that far outnumbered acceptances. However, I believe I am who I am today because of the learning and lessons I gained through that process.

While the title 'Work Expert' is a bit embarrassing, I hope my story of constantly challenging myself and learning from those experiences can reach and help those who have similar concerns.

Thank you for taking the time to watch, and it would be a great help if you could share your thoughts about the video.

https://youtu.be/UR9PL1Jz-qs?si=nYUszHC3GrU21K-Q

Thank you to all the students taking this course, and if you encounter any parts where the explanation is insufficient, parts that are difficult to understand, or difficulties you face when applying this to your actual work, please post them as questions and I will respond as quickly as possible.

Thank you

Best regards, Kang Byeongjin

Are you thinking about developing/operating an LLM service? Our company is planning to hold a meetup to share LlamaIndex and LLM development/operation know-how, so we are providing you with the relevant information~

The 2nd GenAI Connect Day, an LLM Workshop for developers who operate or plan to start an LLM service, will be held on September 26th at GS Tower. This event is hosted by GS Group’s Open Innovation Community 52g in collaboration with LlamaIndex and Pie&AI from DeepLearning.AI. LlamaIndex’s AI Engineer Pierre-Loic Doulcet will present on-site, and Liner’s Hoon Heo and AutoRAG researcher Jeffrey Kim will share their know-how and insights related to LLM service development. We hope you will gain a lot of insights from AI engineers in various fields and have networking opportunities at the after-beer party!

[Guide] GenAI Connect Day #2: LlamaIndex x 52g

📅 Date: 6:30 PM, Thursday, September 26, 2024

🏢 Location: Keeeet Open Hall, 25th floor, GS Tower

👍 Target: Developers who are developing services using LLM

🚨 Application Period: 2024. 9. 9 (Mon) ~ 2024. 9. 24 (Tue)

✅ Speaker introduction and application link: https://bit.ly/52g-genai-2

🚀 Inquiry: GS Corp. Manager Byungjin Kang ( jason.kang@gs.co.kr )

Although it is not officially(?) announced, we are planning to prepare various products for developers, so please show a lot of interest!

Hello, this is Kang Byeong-jin.

We have recently released a course on constructing RAG using the popular Large Language Model (LLM).

This is my main job at my company these days, and I've been struggling with this since last year... I prepared this lecture in the hopes that others will have a little less trouble.

The lecture utilizes the OpenAI API, and since it utilizes LangChain, you can easily follow along by using packages that utilize LangChain, such as Claude, Cohere, Upstage, etc. If you look at the lecture content, you can use ChatAnthropic , ChatCohere , ChatUpstage , etc. for the parts corresponding to llm in LCEL, and if you want to utilize Llama3 by utilizing Ollama , you can use ChatOllama .

If you need a lecture using Ollama or want to know more about LLM, please leave a review or question about the lecture or send me an email to jasonkang14@gmail.com . It will be helpful for me to prepare the next lecture!

Hello, this is Kang Byeong-jin.

It's been about a week since I released the course, and I appreciate all the interest you've shown.

2.5 Auto Scaling Group을 활용한 안정적인 서비스 운영 , I explained how to add various alarms. I added a method to utilize Scheduled Action because I thought it would be nice to actually see instances increase and decrease.

As I explain in the lecture, this is a method that is often used 특정한 날에 트래픽이 몰릴 것으로 예상될 때 . For example, when opening a raffle on Friday at 23:00 or when there is an event, it is a method to prepare for the surge in traffic in advance. And how to reduce the increased instances again is also included in the corresponding section.

Some of you may already be familiar with it, but I think it would be helpful to also refer to the added sections, so I'm sharing them with you.

If you have any questions, do not understand, or find something lacking in explanation while taking the class, please post a question! We will update the lecture materials and videos.

thank you