In the past half month, I have become somewhat addicted to Vibe Coding.

It's not that "I want to create an amazing product" kind of obsession, but rather a sudden realization that many small ideas that have been lingering in my mind can actually be brought to life bit by bit by myself.

As everyone knows, Vibe Coding is about using natural language to command AI to write code for you, bringing a product to life.

I usually use Codex and Claude Code together to describe requirements and functionality modules; they help me write code, and when I reach my limit, I switch to CLI to access the DeepSeek API and keep going.

1. Those Thoughts of "Wanting to Do but Never Doing"

In the past, a bunch of ideas would often pop up in my mind.

For example, wouldn't it be great to have a panel where I could view US stocks, crypto, Hong Kong stocks, and A-shares all in one place, instead of switching back and forth between several software every day?

For instance, couldn’t there be a monitoring system that alerts me immediately when a specific asset suddenly rises or falls, and additionally tells me which targets or sectors are related?

Or how about creating an investment map, where while researching a lane, I focus not only on a single project but also lay out the entire network of upstream, downstream, benefiting targets, potential risks, and related assets?

Additionally, since there are many bets in the prediction market (PM) regarding unlisted company valuations, market capitalization overtaking, and macro events, could we somehow display this data alongside news nodes and changes in the secondary market?

There are plenty of ideas, but actually executing them is too troublesome.

You need to understand code, know how to design pages, connect data, and keep modifying; if you hire someone, the cost is high, and the requirements may not be clear. After going back and forth a few times, most ideas end up being dismissed with the phrase, "Forget it, just make do with Excel for now."

However, after experimenting with Vibe Coding for half a month, I realized that things are genuinely different now.

I started building some rough tools that can solve problems; an idea pops up, and I can get it into the system the same day, rather than letting it scatter in chat records, bookmarks, and my mind.

2. The Four Small Tools I Created in Half a Month

In the past half month, I mainly created four things (I won't count other miscellaneous small tools).

The first: Cross-Market Asset Panel

The origin is very simple. My assets are scattered across several places: Hong Kong stocks, US stocks in a brokerage app, crypto on a trading platform, and A-shares in another software.

Every day when I want to take a look at my overall situation, I have to open each one and switch back and forth; after completing the round, I still can't piece together the whole picture. So the first thing I did was put all my holdings into one page:

At the top are the total assets and today's profit and loss; below is divided by market—one column for US stocks, one for crypto, and one for each of Hong Kong stocks and A-shares. A quick glance gives a clear overview of the status of everything at home, who is up and who is down today.

After finishing, I found it quite useful, and I couldn't help but add more tabs one by one, as new needs emerged while using it:

• Abnormal Movement Monitoring: I set up the targets and thresholds I want to focus on in advance, and if any suddenly surges or drops, it directly marks them for me, saving me from constantly watching the market.
• Investment Map: When studying a specific lane, I visualize the upstream, downstream, benefiting targets, risk points, and related assets into a single network, making it easier to trace the financial transmission chain and relationships;
• Memo + Review: Documentation of why I was optimistic, what happened later, and where I was right or wrong, allowing for easy reference later;

This panel contains all my real holdings, making it relatively private, so I deployed it locally.

The second: PM Bet Monitoring

This is specifically for monitoring the prediction market.

To explain simply, a prediction market (like PM) is where people bet real money on whether a future event will occur, with the price itself representing the market's perceived probability— for instance, the "Will SpaceX reach a $2 trillion market cap by the end of June?" bet marked at 0.8 means that the market believes there is an 80% chance it will happen.

The bets I care about include "Will OpenAI/Anthropic see an increase in valuation by the end of the year?" "Will a certain market cap overtaking event occur among the Seven Sisters?" "Will xx and xx meet?" Previously, I had to look them up one by one, but now I have gathered them all into a single dashboard, correlating probability changes with news events and fluctuations in the secondary market, making it clear who moves first and who influences whom.

I also classified these bets according to my own standards (internally I call them T1 (high certainty)/ T2 (relatively stable)/ T3 (pure speculation)), ranking them by expected returns, so at a glance, I can identify which are merely noise.

To be honest, my slight advantage in this market is my access to information in Chinese and the political and economic dynamics in East Asia—since many are dominated by Western players, pricing in this area often lags, and the opportunity lies in that time difference.

The third: Operations Backend

This is unrelated to investing; it's used for writing my own content.

When I select topics, write articles, and publish across several platforms, I rely on my memory and chat records to keep track of progress, which often gets messy. Thus, I created a small backend to manage it, encompassing a topic list, article progress, publication platforms, and an inspiration box.

Since I might need this on the go, I deployed it online instead of locally—using GitHub + Vercel provides easy access and modification on my phone, which is quite convenient.

The fourth: One-Click Formatting Tool

This was mainly created to solve a personal need. After writing an article, I need to format it for many platforms, especially since each Web3 media has different formatting rules, which is quite time-consuming to adjust manually.

Thus, I made a small tool, working with a coded browser Tampermonkey script. You drop in a Markdown or Word original draft, and it automatically converts it into the format for each platform and inserts images accordingly. It's not advanced, but it saves a bit of mechanical work every day.

In fact, these four tools are still quite rudimentary, perhaps even a bit ugly, and can't be described as mature products, but they are already very useful to me because once an idea appears, I can immediately get it into the system, instead of letting it scatter and be forgotten.

This is the change I find most significant.

3. The Way Ordinary People Research Investments Has Really Changed

Because of this, I increasingly feel that ordinary people don't necessarily have to start with overly complicated models when investing, but at least should have a few basic systems of their own.

Because AI's impact on ordinary people is not making you suddenly become a master, but enabling many "things you wanted to do but couldn't" to at least take shape initially.

Especially for someone like me who looks at the market every day, the feeling is particularly significant. As long as there are ideas, each ordinary investor can slowly accumulate a few basic systems of their own:

• Asset Observation System: What assets are you focusing on, which markets do they belong to, and what changes have occurred recently;
• Signal Monitoring System: What events, once they occur, might indicate that market expectations are changing;
• Map Organization System: A lane is not just one point, but a network; who is upstream, who is downstream, who consumes sentiment, who consumes performance, and who consumes capital. Especially in the past year, stocks in the AI sector have been rewarding those who can fully understand a lane (from HPC to optical modules to storage chains);
• Review System: Why were you optimistic at the time, what happened later, and where were you right or wrong;

These things were not impossible to do before; they were just too troublesome and hard to sustain. The greatest significance of AI is that it has eliminated a big chunk of that trouble.

You might not know how to write code, but you can describe your needs and gradually build your own product design, without needing to complete it all at once. First release a version, then modify it while using it.

This is also the aspect of Vibe Coding that attracts me the most—the feedback is so quick. Previously, the time between an idea emerging and its implementation could be long enough for you to forget why you wanted to do it in the first place.

Now, if I think of a function today, I can test it the same day; if I'm not satisfied after testing, I can modify it immediately; if new requirements emerge after two days, I can continue to iterate.

This closed loop of "idea—implementation—usage—feedback—modification" once in motion truly makes it hard to stop.

In Conclusion

This article serves as the first record of the new stage of "Too Happy Tyler."

Moving forward, I will try to update regularly to document my investment thoughts, tools tested, on-chain practices, and arbitrage research, as well as some educational/introductory Web3 practices and investment knowledge points.

Feel free to follow along and reach out anytime.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。