Exclusive Interview with "AI Investment Research Elite Incubator Program" Trainee Little Flower Cat - AI Market Assistant
Written by: TECHUB NEWS Hong Kong Report | Reporter Alma Li | September 2026
Introduction:
From hearing about Bitcoin mining in the early years and missing several opportunities, to formally entering the cryptocurrency market in 2017; from relying on experience to judge cycles, to trying to structure on-chain costs, miner costs, derivatives, and market sentiment into signals, Little Flower Cat is building an AI market research system.
He does not position the system as a "buy and sell button," but hopes it helps users transform scattered information into a reviewable scoring framework, reducing hesitation and fear caused by uncertainty in a highly volatile market. For him, AI is a tool to enhance efficiency and magnify experience. Ultimately, it is still people who define problems, understand risks, and bear the consequences of decisions.
Core Viewpoint:
"AI will not bear the risk for you, but it can help you know what you are actually basing your decisions on before making them."
TECHUB NEWS: Please introduce yourself. When did you start getting in touch with Bitcoin and the Web3 market?
Little Flower Cat: I am in Shanghai, and my daily work is mainly about implementing enterprise digital systems. I have been a product manager in manufacturing factory digital systems for over ten years. I got in touch with Bitcoin quite early; in 2011, a friend asked me to mine, but I didn’t study its value seriously back then, so I missed out. In fact, I had seen related discussions in some computer magazines and overseas forums around 2009, but I didn't understand much at the time.
Little Flower Cat: During the bull market in 2013, when Bitcoin rose to over a thousand dollars, there was a lot of media coverage; later, the market fell, and due to external influences, I also stopped paying close attention. It wasn't until 2015 when a college classmate asked me how to buy digital currency that I started to seriously look into it, studying the Bitcoin white paper, its operating mechanisms, and value logic. I truly entered the industry in 2017. At that time, topics like Baidu's WanKeYun and hardware mining were very popular; I myself studied computer science and liked to tinker with hardware, so I researched mining machines, joined QQ groups, and slowly truly entered this market.
TECHUB NEWS: From "knowing about Bitcoin" to being willing to focus on it as a long-term research subject, what changes occurred in your understanding?
Little Flower Cat: In the early days, people easily regarded it as something like Q coins, thinking it had no value. There was little Chinese material available, and to research it, one had to look at English material in overseas forums; synchronizing nodes and using wallets were also much more difficult than now. Many people, even if they got Bitcoin, might not know how to store it.
Little Flower Cat: Later, I gradually understood that Bitcoin did not appear suddenly. Prior to that, crypto punks and the digital currency field had already done a lot of exploration; only when conditions such as cryptography and infrastructure matured, along with the background of the financial crisis, did Bitcoin really have the opportunity to emerge. It was only after I understood this history and mechanism that I began to see it as a direction worth researching in the long term.
TECHUB NEWS: After entering the market in 2017, why did you not choose a high-risk, high-position approach?
Little Flower Cat: My personal risk preference is quite conservative. If I think a risk exceeds 50%, I usually don’t want to take it. So I mainly focus on mainstream assets like BTC and ETH; I might buy some altcoins, but it is more about trial and error.
Little Flower Cat: I also have regrets, such as buying BNB at a very low price but not holding on to it. However, I don’t think that means one should abandon risk control. Mainstream currencies may not bring the highest returns, but they are relatively safer. For me, staying in the market is more important than chasing every peak return.
Market Sensitivity: Long-term relies on understanding, short-term relies on cycle judgment
TECHUB NEWS: You want to convert your market experience into an AI market system. In your view, what exactly is "market sensitivity"?
Little Flower Cat: We first need to distinguish between long-term investment and short-term trading. The core of long-term investment is the recognition of asset value and using spare money. I have friends who bought Bitcoin early, put it in cold wallets, and didn't look at it for years, ultimately obtaining high returns. Either they had a strong belief in Bitcoin, or the money they invested was something they could afford to lose, so they were not forced to sell due to short-term fluctuations.
Little Flower Cat: Short-term trading, on the other hand, is more about judgment of market conditions and cycles. Bitcoin has shown clear cyclical characteristics in the past; investors need to identify tops, bottoms, and the stage of the market. If you want to sell high and buy low but your judgment of the cycle is not accurate enough, it is very easy to miss out or get thrown off during extreme volatility.
TECHUB NEWS: What is your view on the recent popular Meme coins or high-volatility assets? Can AI find the next opportunity for investors?
Little Flower Cat: I regard Meme coins as lottery tickets. There are often stories in the market of people making huge multiples, but this easily creates survivor bias: with so many people participating, only a few will actually make money. Not to mention the losses due to trading fees, Gas fees, and platform fees in frequent trading.
Little Flower Cat: One cannot think of it as a path that can easily be replicated just because they see a few successful cases. Investment must have its own system. My view has always been that wealth is the monetization of knowledge. Just because you make money in one field does not mean you can continue to earn in another market that you do not understand; without corresponding knowledge, you may end up losing back your gains.
Turning Experience into System: Using Scoring to Combat Information Fragmentation and Decision Fear
TECHUB NEWS: What made you think of creating this AI market assistant? What problem is it primarily trying to solve?
Little Flower Cat: In the past, I relied mainly on my own experience to judge cycle markets. But the crypto market is very volatile: today you might feel like it’s a low point, but after buying, it could drop another 10% tomorrow or suddenly face a black swan event. Even if you make judgments based on indicators that suggest an appropriate buying position, when it comes time to make a heavy investment decision, fear still sets in.
Little Flower Cat: This fear largely stems from the fact that information has not been quantified. In the past, I would look at data in WeChat groups, X platform, and various websites; everyone has their own preferred indicators, but the information is fragmented, making it impossible to form a comprehensive judgment, relying ultimately on intuition. So, I thought, if I could collect important indicators, set weights, and form explicit scores, at least it could make judgments more quantifiable, reducing hesitation. It’s a bit like a tool or friend that helps you make up your mind, but it doesn’t decide for you.
TECHUB NEWS: What data does the system primarily refer to? Do the judgments for tops and bottoms use the same logic?
Little Flower Cat: No, they are not the same. For bottom judgments, on-chain costs are a relatively core indicator, and miner costs have also been quite effective in the past few cycles; the 200-week moving average or weekly charts will also have a certain weight. Top judgments, however, rely more on derivatives data and market sentiment.
Little Flower Cat: The system does not just rely on one indicator; it combines multiple indicators. The weights are not固定的. An indicator that has been effective in the past does not mean that it holds the same importance in all stages, so we need to continuously adjust them based on changes in market structure.
TECHUB NEWS: How do you design the scoring mechanism of the system?
Little Flower Cat: The system scores daily, but it is not simply a unidirectional score. I designed a "Bull Top Score" and a "Bear Bottom Score," which can be understood as two sets of evidence, bullish and bearish, engaging in contention, like both the affirmative and negative sides presenting arguments at the same time.
Little Flower Cat: Finally, based on the difference in the two scores, the system determines whether the market leans towards the top, the bottom, or is neutral. This way, the system won't give an absolute conclusion just because one indicator is strongly bullish or bearish; when the evidence from both sides is close, the system's "neutral" state itself is also a conclusion, indicating that the current direction is still unclear.
The Market Structure Has Changed: Indicator Weights Must Be Iterated Continuously
TECHUB NEWS: This round of the market is more institutionalized, and macro factors are more complex. How do we confirm that previously effective cycle indicators are still applicable today?
Little Flower Cat: We cannot say that historical indicators are definitely ineffective, nor can we mechanically copy them. For example, while on-chain costs still have reference significance, the judgment of a low point itself cannot be precise to a certain absolute price. If the system indicates that we are entering a bottom stage, I think a more appropriate approach is often to invest in batches and dollar cost average, rather than trying to get in at the lowest point all at once.
Little Flower Cat: Miner costs are similar. In the past, they were relatively important, but now the influence of new mining output on the overall market has decreased, and weight can be lowered. At the same time, changes such as institutional capital entering must also be included in the new weight system. The market has changed, so the system's weights must change accordingly.
TECHUB NEWS: How does the system validate itself? What are the current backtesting results?
Little Flower Cat: I backtested data from 2018 to present. According to the system’s own classification, the accuracy for bottom-oriented judgments is about 90%, and for top-oriented is about 70%. When the system was operational for over twenty days, it also provided judgments for two turning points for tops and bottoms.
Little Flower Cat: However, I must emphasize that these are internal historical backtests and phase observations of the system; they do not represent the future and do not constitute any return guarantees. The market environment will change, and the system must continue to iterate and verify.
TECHUB NEWS: What role does confidence play in the system?
Little Flower Cat: I later added a confidence mechanism. When multiple indicators resonate, the system’s confidence becomes higher; if there is no resonance between indicators, confidence declines.
Little Flower Cat: This does not mean that high confidence is necessarily correct; instead, it lets users know whether the signals behind the current conclusion are consistent. It adds an additional layer of auxiliary judgment, reminding us not to look only at one result, but to also consider the evidence underlying that result.
AI is an Amplifier, Not a Decision-Making Machine
TECHUB NEWS: What was the conclusion the system provided about market conditions at the time of the interview?
Little Flower Cat: At that time, the system judged that the state was neutral, but with slightly higher scores related to the top, indicating a slight top preference within the neutral state. Prior to that, when the market was around $82,000, the system had shown a slight top preference; after a subsequent decline, the state returned to neutral.
Little Flower Cat: For specific points, I can only say that was based on system and personal observation at the time and should not be interpreted as investment advice. If the market experiences a strong macro shock or super black swan event, the trend may change completely. The more important role of the system is to identify phase tendencies, rather than accurately predicting every price.
TECHUB NEWS: Does the system use paid data or free data? What role does AI play in it?
Little Flower Cat: Basic data like K-lines mainly come from publicly available data sources, and other indicators are also obtained through public website information. The underlying development uses AI tools like GPT and Codex.
Little Flower Cat: However, I believe AI itself is not the core barrier. People may use similar models and data, yet the results differ; the difference lies in how you define the problem, what indicators you choose, how you set weights, and how you understand market changes. AI is more like an amplifier of capability: it can enhance development and iteration efficiency, but cannot automatically form correct investment research logic for you.
TECHUB NEWS: Will this system be open to individual investors in the future?
Little Flower Cat: Currently, I lean more towards using it myself or collaborating with institutions. The cognition, risk tolerance, and usage of individual investors vary greatly. If the system is simply packaged as buy and sell signals, it can easily be misused.
Little Flower Cat: Based on the system's backtesting and strategic assumptions, if it can capture several stage fluctuations over a year, it may have an annualized target range of about 20% to 30%; but this is only internal backtesting and strategic assumptions, not actual performance, nor does it constitute any investment advice. Before commercialization, it is more important to continue verifying the system, clarify applicable boundaries, and prioritize risk control.
From Incubator to Product Implementation: Don’t Just Leave Behind PPTs
TECHUB NEWS: What gains has the AI Investment Research Elite Incubator Program brought you?
Little Flower Cat: The incubator has made me refocus on issues like data reliability, future functions, and validation methods. Building an investment research system cannot just look at whether the backtest curve is beautiful, but must confirm whether there was an use of future data that could not have been obtained at that time, the quality of the data, and whether the strategy is still effective after changes in market structure.
Little Flower Cat: In terms of development efficiency, AI tools indeed helped me quickly create many systems and prototypes. But I care more about turning them into works that can be validated and iterated upon, rather than remaining at the idea stage.
TECHUB NEWS: What suggestions do you have for future incubators?
Little Flower Cat: I think the most important thing is to encourage everyone to actually bring their projects to fruition. Many projects tend to stay at the PPT or conceptual stage, but only by entering real usage scenarios will they encounter issues of data, logic, user experience, and risk control, thus having the opportunity for continuous iteration.
Little Flower Cat: Additionally, AI applications should not be concentrated solely in finance. Traditional industries like manufacturing, healthcare, and enterprise digitization also have many worthwhile scenarios to explore. I hope to see more cross-industry projects and discussions in the future.
Editor's Note: This article is based on the audio recording, meeting minutes, and timeline texts from the interview with Little Flower Cat conducted by TECHUB NEWS, with some edits. The content regarding his personal experiences, system functions, data sources, backtesting performances, market judgments, and commercialization plans are based on the interviewee's statements and provided materials; the content involving historical backtests, accuracy rates, target return ranges, and market views has not yet been independently audited, verified, or reviewed by third parties.
Disclaimer: This article is for information exchange and discussion of research methods only and does not constitute any investment advice, solicitation for investment, or recommendations for buying or selling any assets. The prices of securities, futures contracts, and virtual assets can rise or fall, and virtual assets are considered high-risk investments, where investors may lose part or all of their principal. Past performance, historical backtesting, and the interviewee's opinions do not represent future results. Readers should not rely solely on the content of this article to make investment decisions and should carefully assess their own investment goals, financial situations, and risk tolerances, consulting independent professional advice if necessary.
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。
