P young player talks about AI token investment methodology: narrative judgment, buying and selling points

Editor | Wu Shuo Blockchain

This interview deeply explores the investment trends, market status and future opportunities of AI Agent projects. Analyst defioasis shared his investment experience in on-chain assets, memecoin and AI-related projects, including the rise of key projects (such as GOAT, WorldCoin, Turbo, Pippin, etc.) and the market logic behind it. In addition, investment methodology, position management strategies, signs of cooling of market bubbles, as well as changes in industry trends and the importance of project narratives are also analyzed.

Please briefly introduce yourself, and review when you started studying AI Agents, when did you officially start investing money, and what is the current overall return rate

defioasis: Hello everyone, I am defioasis. I am usually interested in the assets on the chain and the derived gameplay. Since last year, the main focus has been on playing things on the chain.

In fact, the research on AI Agents also comes from mining on the chain. I have been observing on-chain assets last year, but that kind of pure Memecoin is not that good. It mainly looks more and does not particularly convince myself to play with the big money.

The turning point was mainly in the late October and early November last year, when GOAT soared to a market value of hundreds of millions of dollars in a short time and Binance's launch of GOAT futures made me rethink this track. Of course, before talking about it, I still want to talk about some AI-related matters before GOAT. Before GOAT, there were two major events. One was WorldCoin. Because Lianchuang was the founder of OpenAI, WLD was regarded as OpenAI's Meme in Crypto, and FDV once exceeded US$100 billion; the second It is Turbo, which claims to be an AI Meme created by GPT, and it is also the target that rose by 200-300 times in CEX last year. So these two things show that everyone is still very interested in the theme of AI+Crypto and is worthy of hype.

Back in the end of October last year, after GOAT got on Binance Futures, I quickly thought of WorldCoin and Turbo, and Binance had the ability to lead the track. At that time, there were 4 assets that I found very eye-catching, two of which were AI-related, one was ai16z and the other was ACT; the other two non-AI-related were LUCE and BAN. Now, ai16z seems to be much better than ACT, but at that time, the founder of ai16z Shaw still had nobody, and ai16z was launched with daosfun, and there was a problem with token contracts being minted, and the fluctuations were particularly large at that time because the pool was very Small. I built a position in ai16z at a market value of US$25 million, which cost about US$3,000, but the pool was very small and the market value was once broken by Fud, and I didn't dare to increase my position. So I finally chose ACT, which is more stable and has a larger number of holders. ACT can be regarded as my first operation to officially open a position in AI investment. The first transaction was to buy the first ACT of $3,000 on November 5. Later, I found that there was a Bitget, and it also bought more than $10,000 ACT at a market value of $22 million, with an average price of $0.022. But at that time, I didn’t expect that I had bought enough positions one after another, and then I was surprised to get on Binance. I remember that I was still attending the conference in Bangkok, but I was actually quite shocked. After joining Binance, the highest market value reached between $700 million and $800 million, but I actually sold most of the market value at $500 million the next morning after Binance news, and the rest have not been sold yet. .

It was also observed later that AI Agents is the only track that goes from universal PumpFun to verticalization and forms scale, and later it goes deep into this track to invest. After ACT, I caught some pretty good targets, such as Pippin, which also had a heavy position and achieved a return of more than ten times. The overall position used to invest in AI alone actually reached 7-8 times since November, and has recently retreated a little, about 5 times.

What is your approach to researching AI Agent projects? Can you give a case of the most profound AI Agent project you have studied so that more people can understand your research methodology

defioasis: The current AI Agent or some assets related to AI are very different from previous AI projects. Now, what is the basic thing about fair launch assets based on PumpFun-like launches, so the project party or founder may not have much assets in their hands, and may be less than some snipers or big players. This means that the founder's character, philosophy and background are very important. Otherwise, if it may not have morality or is not a real name, it can abandon the market and open a new market at any time. So my methodology for AI Agent is first based on people, whether he really wants to do it, whether he has the ability to do it, and what he can do it.

Let me give you an example, taking the experience of the Pippin project as an example. Originally known to Yohei Nakajima (Yohei Nakashima) from Solana AI Hakathon’s judges on December 11, Solana AI Hakathoh is a hackathon event organized by ai16z and Solana. At that time, when I looked at the list of judges, I found a Japanese man, Yohei Nakajima, the founder of Pippin. It was quite interesting to see him doing a child-oriented AI agents at that time because he had never seen agents of the same type. Another aspect is that as a judge, the judge is obviously more valuable than if you need to participate in the competition. Later, Yohei Nakajima went further to study and found that he himself was the founder of BabyAGI, and there were more than 20,000 stars on BabyAGI's Github. I searched some information at that time and found that BabyAGI is quite powerful. As an AGI concept product, it has been cited by many media papers, which also proves its strength. In addition, Yohei Nakajima himself is a partner of Untapped Capital. The main Web3 projects he invested in are Pixel, and Pixle has joined Binance. Overall, Pippin founder Yohei Nakajima has excellent technology and capital resources. As a person with real name and a certain prestige, the probability of him being a random rug is much smaller.

At that time, Pippin's market value was around $20 million, but there were actually not many people paying attention to it. The market value of US$20 million also meets my own requirements for purchasing asset subjects, because I prefer to buy within the range of 10-20-30M market value. I bought two thousandths of the total amount one after another, and two thousandths is the single maximum position limit I defined, which cost about US$40,000. He actually fell below $10 million later, but I bought the chips at that time. Founder backgrounds and technologies like that will not change with price fluctuations. So after the collection, I didn't care.

Later, Pippin announced that it would transform into a framework, from a single AI Agent to an AI Framework, the valuation was a straight increase. After this news, even though the framework has not yet been developed, everyone's recognition of the founder's technology and capabilities has also soared Pippin to US$300 million. Making a framework means that it is possible to become a split plate. The market pricing of framework is the highest in the Ai agents track at present, and the framework or ecosystem that can form a split plate model has the opportunity to exceed US$1 billion market value.

Which AI Agents projects are you optimistic about? What do these projects do? Why are you optimistic?

defioasis: I am very optimistic, like Pippin, but I won’t talk about it when the market value is relatively high. I usually choose around $20 million, but there are actually fewer ones who will have a heavy position.

Currently, there are two main directions. One is gold hunting in Solana AI Hackers Pine. It has ended now and there are quite a few awards in it. Recently, it is being screened. The other is Virtuals to Solana and working with Jupiter. There are probably many interesting projects later, because Virtuals has proved its success on Base. Of course I'm still watching this.

Here we mainly talk about some projects from Solana Hackers. I'll share one here, but this is not investment advice. AgentiPy, a project I am observing recently, is actually an open source framework, which uses python to connect ai agents to solana on-chain applications. According to the roadmap, it may launch an autonomous narrative trading robot in Q1 and launchpad in Q2; most importantly, it mentioned that its token APY will participate as a flywheel. I studied the economics of APY tokens and designed well. Although it is also based on fair launches like PumpFun, it hit 40% of the chips and put them all in Streamflow for two years to unlock linearly, which shows that the team is still determined. of. AgentiPy's Lianchuang and CTO have both been recommended by Solana. It is at least a kind of endorsement after coming out of the Solana Hackers. Of course, the overall situation is still relatively early, and there is a lot of uncertainty. I will also pay more attention to the projects launched on Virtuals after cross-chaining to Solana.

If we look at it overall, AI is already slowly reaching the stage of AI Application. In addition to framework, AI+ applications will also remain focused on the latter, especially AI+DeFi, which is a combination of AI and Crypto native narrative, DeFi assets and flywheels. There may also be some good opportunities, but it is still in a very early stage. At the stage, I didn’t see any good targets. I still keep an eye on it now and haven't bought any new assets recently.

What do you think about the current AI Agents track and market status? Do you think the heat of AI Agent can last for a long time or do you think this bubble has reached its peak?

defioasis: It has been quite cold lately, but I don’t think it will end yet. AI is still relatively making sense. AI outside the circle is still iterating and developing rapidly, and is in the upward momentum of technology and capital. This is a more important basic market. In fact, many AI targets are actually driven by the outside world, whether it is narrative or talent. Shaw was also a fringe person in web2 back then. Now, ai16z has become the leader in Crypto AI. Influenced by Shaw, I think more technical talents from traditional industries will come in to do AI. Crypto AI itself is significantly behind outside the circle, so any turbulence outside the circle and major updates will be transmitted to Crypto to form a new narrative and new sub-track.

From another perspective, from the perspective of the circle, AI Agents is the only one that has formed a scale vertical track from the universal PumpFun. DeSci may be half a piece, and it has been quite cold recently. Apart from that, no other track can move from general to vertical, which actually shows that there is a strong demand for AI narrative. Now AI Agents are relatively cool, which is to cool down the previous overheated cooling. There is a widespread single agents now, forcing everyone to do frameworks and de-voltages, and everyone is gradually becoming more tired of this kind of thing. Therefore, if AI+ applications, especially AI+Crypto native narrative, can be established, I believe it will re-drive a wave of craze and new opportunities.

Experiences about investment and trading, what are the tips for building positions and exits?

defioasis: The above are actually talking about the issue of investment targets, but I actually think position management is more important. The target selection is nothing more than technology, resources and background. Although it is based on PumpFun for fair launch, the research ideas are not much different from those of VC coins, such as technology, resources, background, team, endorsement, etc., as well as analysis of chip structure. , blue chip address of rat warehouse, etc.

Next, we will focus on the issue of position management. Now a pretty good asset can basically go through three stages: PvP, second stage and upper stage, but most PvPs are gone after the end. I usually mainly play the second stage, and I will focus on the market value range of US$1,000-2,000-30 million, which I also call the on-chain dessert range. Because I found that some pretty good assets will probably consolidate and fluctuate at this position after they are born and rise and fall. I usually only do asset allocation in this range. Those who have experienced a 70% retracement that is more stable at US$1,000-2,000-30 million will pay attention.

When I see a pretty good asset, I will add it to the watchlist, which will be divided into S-level, A-level and others. S-level is a currency that can form a mode disk. For example, it is a mother currency ecosystem that can continuously go from scratch Profits from the output sub-coins, for example, as a transaction-to-asset, the sub-coins create wealth effects, and this mother coin is required to trade to generate constant demand. This is the same as the SOL of PumpFun and the Flywheel of Virtuals mother coin. of. Therefore, most S-classes are frameworks, and there is a chance to form ecological or model flywheels. Why Pippin can rise is because the market has high expectations for it to make a framework.

Level A is mainly judged from the narrative, background, technology, resources and team (which cannot form a pattern). The past background of the founder and founding team is very important, such as it is the background of the Solana Foundation, a lot of Github stars, has made products that have attracted attention, and is a good narrative that may develop into a sub-track, etc.

Generally speaking, both A and S levels will build positions, and each position is about 2-3000U. However, I will be stricter about whether to increase my position or to increase my position. I need to observe for a while and decide whether to increase my position. I mainly look at what the community and developers are doing in daily life.

I also have certain standards for the amount of single coin held by each coin. The maximum holding of a single currency is 0.2%, which means that the total cost of buying around the market capitalization of US$20 million is about 40,000 U.S. Every time you choose to increase your position, think about the benchmark. If you judge that the target is contrary to the initial buying logic, you will decisively give up the increase in your position, and the remaining positions will be temporarily reserved to be determined; if it falls sharply below the box market value range, you will decisively give up the increase in your position, and the rest Positions are temporarily retained. In principle, if you get at least 10 times, then consider whether to sell it. It will be considered within the market value range of about 200-300m.

The targets for building positions are often to ensure that they are cautious in the car, and to gradually increase positions in batches and set up a thousand-2 hard tops to avoid adding positions all the way to make funds trapped on a certain target due to excessive blind confidence. Of course, as long as you choose to increase your position, you will believe in your judgment on this target.

All the above are some second-section gameplay. I feel that in the market, both users, Devs and coin issuing factories seem to be more familiar with the routine of these second sections, and I feel that the second section is still relatively difficult to play recently. So I am also doing some lottery streaming gameplay, which is PvP. At this time, it is no longer limited to AI. I have prepared a few SOLs, and I will buy at 0.1 SOL each time, keep an eye on the market, monitor and follow some orders to select addresses, and make a big deal with small ones. I have felt in the past two days that in this market, the lottery streaming game is easier to make money. This may be a probability. As long as there are so many plates, as long as the investment is small enough and scattered, and some high winning rate addresses are monitored as an auxiliary, the lottery stream can recover all the previous losses by just one shot. In this lack of a big market, it will It is a good choice, and to some extent it is also to prepare for potential big markets. It is very important to always maintain a profitable feeling and mentality.