MEXC Product Director Vivien Lin said the exchange’s product philosophy is still heavily shaped by her years in traditional finance, where she worked on derivatives at institutions including Morgan Stanley and Deutsche Bank. In an interview with BlockTempo on Aug. 21 at Melasti Beach in Bali, Lin discussed how that background now influences MEXC’s approach to risk controls, user segmentation, stock products, and AI tools.
The article was presented as sponsored content written and provided by MEXC. BlockTempo said the piece does not represent its editorial position and should not be treated as investment, buy, or sell advice.
From Wall Street derivatives to crypto product design
Lin said she used to view her time in traditional finance as routine and repetitive, but now sees that period as the source of many of her current product ideas. Her earlier work focused on equity and foreign-exchange derivatives, especially option-linked products, where risk management often meant dealing with Greek exposures and positions that could not be fully hedged in the market.
That experience, she said, forced her to think clearly about what kinds of risk she was willing to take, what kinds she could take, and what to do when markets moved away from the original thesis. She described that habit of planning ahead as something that has become deeply ingrained and still guides her work at MEXC.
Why exchange rules start with risk
Asked whether crypto users face much higher risk than institutional market participants, Lin said the answer is yes, and that this directly changes how products are configured. From an exchange operator’s perspective, she said, the biggest danger is bankruptcy risk tied to highly leveraged positions during violent market moves. In those cases, a user can lose more than the funds in the account, and the loss can spill over to the platform.
She said MEXC wants to avoid both platform-level loss and user liquidation, which is why risk controls are built into trading parameters. One example is the maximum order size on each trade, such as a cap on how many BTC can be traded in a single order. Lin said that limit is not there to stop people from trading freely; it reflects available liquidity, because a large market order can produce wide spreads and heavy slippage.
Leverage settings work in a similar way. Lin said the first factor is the user’s trading experience and risk tolerance, and the second is market liquidity. If an asset has thinner liquidity, then even when high leverage is available, the position size allowed under that leverage will be smaller. She said these settings may look isolated on the surface, but they are all connected through a top-down framework that determines which assets are available, which tools can be used, and which user groups should have access to them.
Psychology and product experience
Lin also said her background in psychology affects how she thinks about product design. Trading, in her words, is a serious process filled with price swings, but the product manager’s job is to create a more relaxed and intuitive experience because user experience is also a psychological and emotional experience.
She gave a liquidation-risk example. If a user is close to being liquidated, she said, that is not the time to push a campaign notification. If the platform is going to interrupt the user with a pop-up, it should send the person straight to a quick deposit path so margin can be topped up immediately. In her view, that is what actually solves the user’s most urgent problem at that moment.
AI makes finer user segmentation workable
Lin said exchanges have long grouped users into broad buckets such as beginners, VIPs, and institutions, but those labels are often too coarse. Even among beginners, she noted, there are major differences by age and behavior. The same applies to VIP users: some are asset allocators, while others are active traders.
She used a simple comparison. An asset-focused VIP who has never experienced liquidation may become extremely panicked when a position approaches the liquidation threshold. A trader who uses 50x leverage on meme tokens all the time is likely to react very differently. Lin said those two people should not receive exactly the same product experience under the same market condition.
According to her, the idea itself is not new. The hard part used to be execution. It was difficult to analyze something like 32 large user categories across 100 scenarios when teams had limited staffing. AI has changed that, she said. If the data is clean and the event tracking is done properly in the data pipeline, segmentation and matching can move from theory into a workable methodology and a repeatable SOP.
Why MEXC chose a real-share route for RealStocks
Lin spent part of the conversation explaining why MEXC launched RealStocks as a real shareholding service through brokers, rather than relying only on tokenized stock products like those offered elsewhere. She said the decision came back to a user-centric question: what do the users MEXC wants to attract actually need?
For MEXC, one major goal of building out traditional finance products is to reach across market circles and bring traditional finance users onto the platform. Research showed that this audience cares deeply about whether there is a real asset behind the product and whether the exposure is merely a paper claim.
Lin said liquidity was another issue. U.S. equities have a mature and deep market structure in their native form, but once those assets are tokenized, the market-making ecosystem becomes less developed. Traditional finance market makers still see tokenized instruments as new, she said, and many crypto-native market makers also view them as unfamiliar instruments. If a product cannot clearly demonstrate that it represents real stock ownership with dividends and shareholder rights, and if liquidity is also weaker, then it becomes harder to compete.
That, she said, is why MEXC chose what she called a harder but more correct path, one aimed directly at the trading format traditional finance users already trust and understand.
She also made clear that RealStocks does not replace MEXC’s tokenized U.S. equity offering. MEXC joined the Ondo Global Markets Alliance in September 2025, listed tokenized U.S. stocks issued by Ondo, and has continued expanding that lineup. RealStocks is an additional access route for real shares, she said, not a substitute.
Voting rights are theoretically supported, but not public yet
When asked whether holders of RealStocks would have voting rights, Lin said the product infrastructure already supports that in principle, and that from a legal standpoint it is also theoretically supported. Still, MEXC has not made the feature public.
Her reason was product complexity. Corporate actions such as stock splits, reverse splits, consolidations, and shareholder voting all involve highly individualized and often ad hoc design work. At an early stage of market exploration, she said, introducing that level of complexity too soon could make the product too heavy. MEXC is taking a step-by-step approach and will move when user demand becomes clear.
On the broader U.S. stock topic, Lin also mentioned SpaceX. She said MEXC was fortunate to secure an allocation and, because the asset could be obtained at the original price, the team wanted to pass that profit opportunity on to loyal trading users.
AI Copilot changes the information each user sees
Lin said MEXC’s AI tools already customize information according to user behavior, even if the interface still looks the same on the surface in the web version or the app. As users interact with the platform, they generate data that the exchange’s AI copilot can observe. From there, it can infer traits such as whether someone is risk-averse or more inclined toward yield-style products.
She said that if the system thinks a user is likely to be a beginner and that person asks, “What do you think about BTC today?”, the answer may be framed in plain language: a news event involving Trump, the rough trading range, and a simple description of what is happening in the market. If another user is constantly trading with high leverage and uses more advanced tactics, MEXC would classify that person as experienced. The answer to the exact same BTC question could then move straight to open interest, funding rate, market sentiment, prediction market interpretation, and market microstructure.
On the surface, Lin said, both users are entering the same place. In practice, they are stepping into two different worlds.
Why there is a one-click order button next to news
Lin said the reason MEXC places a one-click order button beside AI News Radar is simple: user intent. If someone is reading news carefully, that person may already be preparing to trade. Product design, in her view, should shorten the distance between the page the user is on and the page the user wants to reach next.
Traditional finance users face the bigger learning curve
The interview also touched on a joint MEXC-CoinGecko report showing that 61.9% of crypto-native users have already started trading traditional assets on crypto exchanges. Lin said the harder migration goes in the other direction: bringing traditional finance users into crypto.
She argued that the more experience someone has in traditional finance, the stronger the first culture shock can be. Part of that shock comes from product mechanics. In traditional markets, spot is spot and derivatives are understood in a familiar way, often with delivery-based structures. In crypto, by contrast, users have to deal with mechanisms such as funding rate, which constantly reflects the pricing gap between spot and perpetuals through detailed formulas rather than a simple comparison of two last-traded prices.
Lin said traditional finance users are often willing to accept uncertainty from the market itself, but not uncertainty embedded in the mechanism. They can live with prices moving against them. What they do not like is finding out there was a rule they did not fully understand or that the product worked differently from what they expected.
She said many people she knows who moved from traditional finance into crypto trading were initially overwhelmed by the number of parameters involved: lending rate, funding rate, unified margin system, borrow rate, and various borrowing limits. Their concern is not only market risk but the fear of triggering some hidden mechanism that results in liquidation.
Still, once that hurdle is cleared, Lin said users often realize that crypto products are remarkably smooth, and that the industry is highly transparent because the rules are generally spelled out in documents. From her own experience building products on both sides, she said crypto has already produced very strong user experiences. The next task is to present complicated documentation in a much simpler way.
Her current favorite product is AI Strategy
Asked which recent product makes her the proudest, Lin named AI Strategy without hesitation.
She said the front-end experience is relatively simple. Users adjust a few parameters, confirm, and the strategy runs automatically. Behind that simple flow, though, MEXC has invested more effort than it did in any of its earlier grid products. The team drew from classic quantitative strategies used by top traditional finance traders, then discovered that those models did not fit the crypto market perfectly and had to be reworked heavily.
Lin said users may open the page and feel surprised by how well the earlier batch of AI strategies performed, but she stressed that the results were not luck. She added that while many of MEXC’s previous products were built mainly for retail beginners, the company has started shifting its focus toward professional users and users coming from traditional finance.
Behavior can reveal what self-image misses
At the end of the interview, Lin was asked about a common line from professional traders: learning to trade starts with knowing yourself. Her answer drew a distinction between psychology and behavior. From the perspective of self-awareness, she said, a person naturally knows themselves best. From the perspective of behavior, however, actions are often the truest expression of who that person is.
She said that if AI judges her to be highly risk-seeking while she thinks of herself as conservative, the useful next step is not to reject the AI result outright. It is to ask where the mismatch appears. In which situations is she conservative, and in which ones does she become aggressive? Did losses make her more aggressive? Did FOMO do it? Serious traders, she said, have to work through both self-perception and behavioral recognition.
Lin linked that view to her own career. Because she began as a trader, she became used to having every decision examined closely. In traditional finance, she said, risk managers calculated her risk every day, and when she lost money, senior managers would ask why she went long in one place and short in another. That environment trained her to explain her behavior with precision.
She gave a BTC example. If she sees bitcoin rising quickly and feels the urge to chase the move, she tells herself that the impulse may be pure FOMO rather than analysis. Once she sees that clearly, she cuts the size of the trade instead of going all in.
Her practical advice is to decide in advance where position sizing has to stop, whether a trader is naturally aggressive or simply feels unusually confident in a specific setup. One signal matters more than people think, she said: if a position keeps waking you up in the middle of the night to check your phone, the position is too big and needs to be reduced. The body will tell you when you are not comfortable, even if your mind keeps saying you can handle more.
Her rule for trading over the long term was brief: make sure you can sleep happily.