Bojie Li

Switching from IDE and Vim to Cursor

Previously, I used JetBrains series IDEs (PyCharm, CLion) for larger projects and vim for smaller ones. The most annoying part of developing larger projects is writing boilerplate code. Most of the time is not spent thinking about the design of functionalities or algorithms but on boilerplate code.

Cursor is an AI-assisted programming IDE similar to GitHub co-pilot, with an interface quite similar to VS Code. When Cursor was first open-sourced in 2023, I started using it, but it wasn’t particularly useful due to the limitations of the foundational model at that time. After GPT-4o was released in May this year, I started using Cursor again and found it more convenient than asking code questions in ChatGPT. Firstly, there is no need to switch windows back and forth, and secondly, Cursor has context, making queries more efficient.

In the past three months, with the more powerful coding capabilities of Claude 3.5 Sonnet, I have completely switched from PyCharm and Vim to Cursor because Cursor’s development efficiency is much higher than PyCharm with AI completion features, doubling the overall development efficiency. My GitHub has also easily stayed all green in the past three months.

Cursor can help quickly get started with new languages and frameworks

Cursor is not only useful for improving development efficiency but also for quickly familiarizing ourselves with new programming languages, frameworks, and tech stacks. For example, writing backends in Go, frontends in React, and smart contracts in Solidity were all new to me, but with AI-assisted programming, these are not difficult. If I had such powerful AI when I was in school, I could have learned many more programming skills.

Rumors about OpenAI o1 started with last year’s Q*, and this year’s Strawberry fueled the fire again. Apart from the name o1, most of the content has already been speculated: using reinforcement learning methods to teach large models more efficient Chain-of-Thought thinking, significantly enhancing the model’s reasoning ability.

I won’t repeat OpenAI’s official test data here. From my experience, the overall effect is very good, and the claims are not exaggerated.

• It can score over 120 points on the 2024 college entrance exam math paper (out of 150) and completed the test in just 10 minutes.
• It can solve elementary school math competition problems correctly, thinking of both standard equation methods and “clever solutions” suitable for elementary students.
• It can solve previously challenging problems for large models, such as determining whether 3.8 or 3.11 is larger, whether Pi or 3.1416 is larger, and how many r’s are in “strawberry.”
• In programming, it can independently complete the development of a demo project, seemingly stronger in coding ability than the current best, Claude 3.5 Sonnet.
• An example in the OpenAI o1 System Card shows that when solving a CTF problem, the remote verification environment’s container broke, and o1-preview found a vulnerability in the competition platform, started a new container, and directly read the flag. Although OpenAI intended to highlight AI’s security risks, this also demonstrates o1’s ability to actively interact with the environment to solve problems.

Some say that OpenAI has created such a powerful model that the gap with other companies has widened, making small companies unnecessary. I believe the situation is quite the opposite. For AI companies and academia without the capability to train foundational models themselves, as well as AI Infra companies and AI Agent companies, this is optimistic news.

The Qixi gift I made for my wife: an AI-generated video composed of 25 AI-generated 5-second clips and a piece of AI-generated music. Most of these videos were created using our static photos combined with textual descriptions of actions, some of which are quirky movements; some were generated by overlaying our photos onto other landscape pictures.

The cost of generating the video was about 10 dollars. Although the result is not as good as Sora and has many obvious physical inaccuracies, it is much better than last year’s open-source models like Stable Video Diffusion, and the consistency with the reference images has also improved.

(Video 02:02, 44 MB)

The two biggest advantages of Web3 are tokenomics and trust. Tokenomics addresses the issue of profit distribution. This article mainly discusses the issue of trust.

The essence of traditional Web2 trust is trust in people. I dare to store my data with Apple and Google because I believe they won’t sell my data. I dare to anonymously complain about my company on platforms like Maimai because I trust they won’t leak my identity. But obviously, in the face of profit, people are not always trustworthy.

How can Web3 better solve the trust issue? I believe that Web3’s trust comes from three main sources: Cryptographic Trust, Decentralized Trust, and Economic Trust.

The essence of cryptographic trust is trusting math, the essence of decentralized trust is trusting that the majority won’t collude to do evil, and the essence of economic trust is trusting that the majority won’t engage in unprofitable trades. Therefore, these three types of trust decrease in reliability.

So why not just use cryptographic trust? Because many problems cannot be solved by cryptographic trust alone. Although these three types of trust decrease in reliability, their application scope increases.

Next, we will introduce these three types of trust one by one.

Cryptographic Trust

• How can I prove my identity without revealing who I am? For example, Maimai needs to verify that I am a member of a certain company, but I don’t want to reveal my exact identity to Maimai. Is this possible?
• How can online games with randomness ensure fairness? For example, how can a Texas Hold’em platform prove that its dealing is absolutely fair and that the dealer is not secretly looking at the cards?

Space Exploration Requires Too Much Fuel

When I was a child, there was an old man in our yard who worked in aerospace. He often explained some aerospace knowledge to me. What impressed me the most was a solar system space map on his wall, similar to the one below.

The old man told me, Doesn’t this solar system space map look a lot like a train route map? But the numbers on it are not distances, but changes in speed (Delta-V).

When he was young, he also hoped to build an extensive space network like a train network, but the stations would no longer be Beijing West, Shanghai Hongqiao, but the Earth’s surface, low Earth orbit, Earth-Moon transfer orbit, Mars transfer orbit, Mars surface, etc. Unfortunately, to date, humans have not visited most of the stations on this map.

The most important reason for this is that human energy technology is too backward in the face of space. Current rockets rely on ejecting propellant for propulsion, and the fastest ejection speed today is only about 4500 meters per second. This is much faster than a bullet, but still not very fast in space. For example, the first cosmic velocity is 7900 meters per second, and considering air resistance and gravity at a 250-kilometer orbit, a speed increment of about 9200 meters per second is needed to enter Earth’s orbit at 250 kilometers high.

More critically, the mass of fuel required by rockets grows exponentially with the required speed change. When I was in elementary school, I didn’t understand this. If the ejection speed of the propellant doesn’t change, shouldn’t the speed be proportional to the amount of fuel burned? For example, if a car’s fuel tank has twice as much fuel, it can travel twice as far.

Nearly 5000 people attended the USTC 2024 Alumni Reunion, and about a quarter of our 2010 class from the School of the Gifted Young returned.

Respected leaders, teachers, and dear alumni:

Good afternoon, everyone! I am Li Bojie from Class 00 of the 2010 cohort. It is a great honor to speak as an alumni representative. In the blink of an eye, it has been ten years since we graduated with our bachelor’s degrees.

First of all, I want to express my most sincere gratitude to my alma mater. From 2010 to 2019, from undergraduate to master’s and doctorate, I met a group of outstanding classmates and alumni who are still my best friends and partners in entrepreneurship. During my Ph.D., Professor Tan from my wife’s lab invited me to give an academic report, and that’s how I met my wife.

(This article is my Zhihu answer to “Are highly career-oriented men suitable as life partners?”)

After starting my own business, I met many entrepreneurs, most of whom are highly career-oriented men.

I discovered an interesting phenomenon: These entrepreneurs have a significantly higher single rate compared to their peers. Moreover, the stability of their marriages is also lower than that of their peers.

High Single Rate

In the fields of AI, mobile internet, and Web3, successful co-founders of startups are generally worth at least a small fortune; even those whose startups haven’t succeeded have very impressive resumes, such as graduating from prestigious universities, holding high positions in major companies, and having various titles and awards. They certainly have no trouble finding great partners. So why is the single rate so high and marriage stability so low?

The core reason is that highly career-oriented men spend most of their time and interest on their careers, with relatively little investment in life, relationships, and family.

A month ago, the interview video by Zhihu “New Figures” was finally released. It was my first time participating in such an interview that included aspects of personal life, and it definitely wasn’t a company PR, as the name and products of our company were never mentioned throughout the entire session, and few people even know the real name of our company.

It seems that Zhihu still maintains journalistic integrity, as they did not let me view the video before publishing it; all editing, titles, and voice-overs were done by the Zhihu editors.

(04:16, 215 MB)

Video shooting locations:

• Beijing office
• Home (interview, cooking with my wife, and some photos)
• Shucun Suburban Park (a place where I often run, the flying electric butterfly was made by me in 2017, it got caught in a tree during the shooting, and our very capable photographer climbed up the tree to retrieve it)

Source: Sohu Technology Interview “Is It Difficult to Develop Large Models Domestically? ByteDance, Baidu, and Unicorns Compete Overseas, Who Will Profit First from Over 70 AI Products?”

Produced by | Sohu Technology

Author | Liang Changjun

“Every day from 9 AM to 3 PM, I have meetings with foreign teams for remote development, internal testing, or bug fixing.” Entrepreneur Li Bojie, who is about to launch an AI product overseas, has been exceptionally busy recently.

This is a C-end AI evaluation product that helps users recommend different AI models or products. He hopes to make this product the “TikTok of the large model era.”

More than a year ago, when Li Bojie decided to leave Huawei to start his own business, he aimed to enter the overseas market. At that time, domestic large models were still in the stage of fierce technical competition, but now more and more companies are choosing the same direction as him.

Whether it’s ByteDance, Baidu, Alibaba, or large model unicorns like MiniMax, Dark Side of the Moon, and Zero One Everything, they are all accelerating their overseas expansion to tap into the global market.

Many companies are quietly making a fortune. Sohu Technology has learned that several overseas products have achieved rapid growth in users and revenue, and some have even started to become profitable. Some products have seen a surge in traffic with AI support, and are expected to achieve profits of 70 to 80 million yuan this year.

In the mobile internet era, Chinese companies went overseas and created TikTok. Now everyone is trying to create the TikTok of the AI era. This is a huge opportunity, but also full of challenges.

(This article was first published on Zhihu answer: “How to develop research taste in the field of computer systems?”)

In the blink of an eye, it’s been nearly 10 years since I graduated from USTC. Yesterday, while discussing with my wife the recent developments of our classmates in the USTC systems circle, I realized that research taste is the most critical factor in determining academic outcomes. The second key factor is hands-on ability.

What is research taste? I believe that research taste is about identifying influential future research directions and topics.

Many students are technically strong, meaning they have strong hands-on skills and system implementation abilities, but still fail to produce influential research outcomes. The main reason is poor research taste, choosing research directions that either merely chase trends without original thought or are too niche to attract attention.

PhD Students’ Research Taste Depends on Their Advisors

I believe that research taste initially depends heavily on the advisor, and later on one’s own vision.