Design Self-study Taste's Bible for Practical AI HW Design, Complete Conquest of CNN Operations (Accelerator Practice Using Verilog HDL + FPGA)

Study the basics of AI hardware with Matbi. As a non-memory design expert, it contains essential knowledge for designing good AI HW.

(5.0) 45 reviews

701 learners

Level Intermediate

Course period Unlimited

Verilog HDL
Verilog HDL
FPGA
FPGA
CNN
CNN
Embedded
Embedded
Verilog HDL
Verilog HDL
FPGA
FPGA
CNN
CNN
Embedded
Embedded

Reviews from Early Learners

5.0

5.0

sunny75

100% enrolled

2024/09/01/Sun 12:45 I listened to the lecture well. It's not about explaining the H/W design method, but it explains CNN and all the knowledge I know. I thought H/W was getting faster, but I realized that I need to think a lot to get faster and know a lot about the functions to apply. The practical lectures were also good every time. I watched the lectures quickly, so I think I'll have to listen to the lectures again later. I bought Arty Z7-20. When I have time, I'll try to follow the practical training well. Thank you for the great lecture.

5.0

kyw0531

100% enrolled

This is a true practical, hands-on lecture where you can clearly understand how AI and FPGA perform inference and how they are configured in H/W to operate in a System. It's an incomparable lecture compared to those that only explain theory, truly highly recommended. I feel myself growing step by step through hands-on practice with Xilinx FPGA, and I'm so grateful to be able to learn and practice the concepts of AMBA and Deep Learning. You are like a true mentor. Thank you very much.

5.0

박승현

100% enrolled

I Love Matbi !!!

What you will gain after the course

  • Insights from a working non-memory designer

  • Essential knowledge for AI HW design

  • Experience in designing HW accelerators

The essential bible for AI HW design!
You can build experience that can be used immediately in practice.

Check out the Tasty Universe roadmap for you who dream of becoming a non-memory design expert.
(Click the image to go to the detailed roadmap page.)


Hello!
Design Self-Study's Matbi here.

This is the third lecture following Verilog HDL Season1 and FPGA Season1.
While not mandatory, the content is primarily structured for those who have completed the prerequisite courses to maximize their learning experience.

This course covers AI HW (AI Hardware) design that non-memory designers and hardware engineers must know.
If you're curious about what AI HW design is and how to achieve optimal AI HW design, please pay attention to this course!

Non-memory semiconductor design essential knowledge!

💡 Please check before taking the course!

  • This course does not cover the prerequisite course curriculum. Questions about content covered in the prerequisite courses are handled in those prerequisite courses, so please enroll only if you are prepared 🙂
  • You can check the detailed introduction video and curriculum through the preview video. Please be sure to watch it before enrolling!

Many opportunities in AI HW design are
waiting for you.

The AI HW design field is a field that has been mature for less than 10 years. As such, numerous domestic and international startups are each striving to become the world's best tech companies.

In fact, the demand for AI hardware such as GPUs, AI ASSPs/ASICs (NPUs), and FPGAs is steadily increasing, and it is expected that their future value will continue to rise for more than 10 years. As indicated by the characteristics of being an emerging technology with a market that has high expected demand, I am confident that the prospects for the AI hardware field will remain bright in the future. This also means that great opportunities await those of you in the AI hardware field.

Source: Semiconductor Engineering

If you study AI HW design, a field where the market actively demands specialized technical skills, your market value will also increase.
Even if you're already earning a salary in the top 1% of the industry, there will definitely be opportunities for you to earn an even higher salary.


After completing all of these lectures

You are ready to design good AI HW.
Since this is an emerging field where technological demand has arisen less than 10 years ago, the market is not yet mature. This means that the direction of technology has not been established and technical know-how is not being passed down either.

Having worked on AI HW design for 4 years and used it in actual products, this course contains core theories and practical exercises that can be immediately applied in real-world situations.
I recommend this course to those preparing to enter as AI HW master's/doctoral students or industry practitioners.


See you in the field 🖐

Thank you sincerely for reading this.
If you've read this far, even if you don't take my course,
I'm confident that you will become excellent design engineers.

The choice is yours.
I look forward to meeting you in the field.

Thank you. From Matbi.

🎁 We're giving discount coupons to Design Self-Study Community members! 🥰

Join the Design Self-Study Community and get discount coupons! Join Link (Click)

Recommended for
these people

Who is this course right for?

  • People who have jobs related to AI HW and are preparing for master's or doctoral degrees

  • Someone with basic experience in digital circuit design

  • Anyone interested in AI HW design

Need to know before starting?

  • Verilog HDL

  • FPGA

  • C

Hello
This is semisgdh

Inflearn Verified

4,372

Learners

381

Reviews

1,316

Answers

5.0

Rating

6

Courses

Hello. I am Matbi from Design Self-Study.

Current) I am designing HW IPs at a Global Top5 Fabless company.

I am creating "real-world semiconductor design lectures" that have never existed before and will be helpful in your professional career.

Design Self-Study Roadmap and improve your skills together.

I am running the Design Self-Study Community to communicate with all of you.

(Join the community and receive a 15% discount coupon.)

Let's enjoy studying and grow together!

More

Curriculum

All

25 lectures ∙ (9hr 24min)

Course Materials:

Lecture resources
Published: 
Last updated: 

Reviews

All

45 reviews

5.0

45 reviews

  • leeug135308님의 프로필 이미지
    leeug135308

    Reviews 2

    Average Rating 5.0

    5

    100% enrolled

    As a doctoral student, I was deeply impressed by this. I'm planning to create an NPU as my defense topic, and it seems I'll need to study Training as well! I'll diligently take the next lecture on AXI and become a hardware design expert!

    • semisgdh
      Instructor

      Hello 🙂 Thank you so much for diving this deeply into the material even during your doctoral studies. Since you're aiming to design NPUs and already have a solid foundation, I'm confident you'll achieve great results in your future research. Hearing that you'll continue with the AXI course and grow into a hardware design expert gives me a great sense of fulfillment. If you need anything, please feel free to ask anytime. I'll prepare even more substantial content for the next course! Happy learning and fighting!

  • soi897님의 프로필 이미지
    soi897

    Reviews 6

    Average Rating 5.0

    5

    80% enrolled

    I had an experience accelerating a given model with GPU in the lab. I really wanted to design it myself, but I was at a loss because there were no proper lectures or books that taught me about hardware accelerator design. Then I came across Matbi's AI hardware design lecture. This lecture was very helpful not only for hardware accelerator design methods but also for understanding the basics of deep learning. I recommend this lecture to those who are curious about hardware accelerator design methods and those who want to efficiently study by selecting knowledge that is helpful for hardware design among the vast amount of deep learning knowledge. I'm looking forward to the next lecture!

    • semisgdh
      Instructor

      Thank you so much for your detailed second comment!! It really helps me a lot in making the next lecture. I made it thinking that it was a lecture for beginners in the field. That means there are a lot of things you can do based on this knowledge in the future, right?! There are a lot of materials on the Internet, so if you read them and study them, it will help you improve your skills. Enjoy :)

  • jumg17047841님의 프로필 이미지
    jumg17047841

    Reviews 3

    Average Rating 5.0

    5

    60% enrolled

    Thank you for the great lecture!

    • semisgdh
      Instructor

      Thank you for leaving a review, Yoonmae! You've already passed the halfway point. If you continue steadily like this, you'll be able to grasp the entire flow of AI accelerator practice at a glance. Let's run together until the finish line!

  • mini9136님의 프로필 이미지
    mini9136

    Reviews 4

    Average Rating 5.0

    Edited

    5

    100% enrolled

    I purchased this course a few years ago, but I am only now leaving a review. Based on my experience implementing AI several times, I found the following aspects of this course particularly good: 1. General introduction to AI H/W and its operating principles 2. Learning computational concepts and H/W design using C/Verilog 3. Acquiring CNN knowledge while debugging the provided code Even for someone like me with some design experience, it was difficult to fully understand everything in one go, but I believe the course structure is easy enough to follow if you listen to it slowly multiple times. In particular, I recommend that students preparing for job placement programs take this course as well. I am rooting for both Matbi and all the students to achieve what they desire..!!

    • semisgdh
      Instructor

      Hello, Arc! I am writing to respond to the valuable review you left. Even though some time has passed since your purchase, I am sincerely grateful that you didn't forget and took the time to leave such a thoughtful review. Despite having experience in AI implementation, it makes me very proud to see that you delved deeply into the core principles, covering overall hardware operation, C/Verilog design, and CNN knowledge. As you mentioned, the content is not easy to understand perfectly all at once. However, if you step through debugging the provided code and repeat the process several times, you will eventually internalize the entire hardware design flow as your own. It gives me great strength that you even recommended this to students preparing for job-linked training programs. I sincerely support you and hope that all your future research and design work goes exactly as you wish. Thank you!

  • jinzhen72284님의 프로필 이미지
    jinzhen72284

    Reviews 4

    Average Rating 5.0

    5

    32% enrolled

    This is a great lecture. Thank you.

    • semisgdh
      Instructor

      Hello Jin-wook Kim, thank you for taking the Practical AI HW Design course and leaving such a great review. (A back-to-back review...?!!!) The content isn't easy, so I'm proud to see that you're following along well. I hope you complete the course until the very end and make this knowledge your own asset. Thank you.

semisgdh's other courses

Check out other courses by the instructor!

Similar courses

Explore other courses in the same field!