Is the rise of quantitative skills an "equal rights moment" for investment?

For a long time, quantitative investment has had an extremely high threshold for ordinary investors. To conduct quantitative trading, one had to afford financial terminals that cost tens of thousands to hundreds of thousands of yuan per year, master Python and backtesting frameworks, and finally understand professional content such as factor testing. With layers of thresholds superimposed, quantitative investment has become an exclusive game for institutions and a small number of high-net-worth individuals.

This situation is now changing. Since 2026, with the advent of the AI-Agent era, encapsulating professional capabilities into Skills that can be called by large models has become a trend.

CLS reporters found that on GitHub and SkillHub, a considerable number of developers have shared quantitative investment-related Skills. In the Agent era, users only need to install these Skills on the Agent to easily call some quantitative functions.

It is worth noting that a large number of quantitative investment-related Skills have also been launched on "WorkBuddy", an office Agent with high daily active users. Reporters found that these Skills have formed a layered matrix from underlying data to upper-level strategies, covering capabilities such as data infrastructure, factor research, portfolio optimization, stock selection application, and backtesting for regular investment, forming a complete "Data - Research - Decision - Execution" pipeline.

As a result, "quantitative equalization" has been pushed to the forefront once again, and the threshold seems to be disappearing. Many quantitative private fund investment managers and practitioners shared their views in interviews with CLS reporters, stating that quantitative Skills reduce the "tool" threshold rather than the difficulty of "investment"; what is truly equalized is knowledge and tools, not excess returns. This discussion on "quantitative equalization" is far more complex than imagined.

The emergence of Skills: Start quantitative research with "one sentence"

In the WorkBuddy ecosystem, quantitative investment-related Skills have formed a quantitative capability matrix, which can be clearly divided into seven levels by capability.

The bottom layer is the data infrastructure, which is the "raw material layer" for all quantitative analysis. Data source connectors such as Wind Alice, NeoData, iFinD, and Tushare open up market, financial, announcement, and macro data that retail investors previously needed to spend tens of thousands to hundreds of thousands of yuan on financial terminals to obtain, allowing users to access the data at any time via natural language queries.

Further up are the quantitative factor research layer, portfolio optimization and timing layer, stock selection and screening application layer, backtesting and regular investment layer, short-term sentiment layer, and strategy report layer.

At the quantitative factor research layer, these Skills cover high-frequency price-volume factors, machine learning-based stock selection, public opinion text factors, portfolio optimization and other directions, and can reproduce formulas, generate codes, and assist in designing index enhancement strategies. Some Skills support diagnosing 55 weekly high-frequency price-volume factor families using indicators such as IC, RANK_IC, ICIR, decile long-short returns and Alpha.

The portfolio, timing and application layers are closer to daily investment. Some Skills try to use machine learning to predict the next-period stock selection capability of different style factors and dynamically adjust the weight of composite factors; functions including full-dimensional A-share stock selection, thematic stock selection, ETF screening, company quality scoring, regular investment backtesting, fund portfolio backtesting, and short-term sentiment main line judgment turn some rules and processes in quantitative research into directly callable tools.

This is also the most easily perceived change in WorkBuddy's quantitative Skills for the market. In the past, when ordinary investors came into contact with quantitative investment, they often only stayed at simple indicators such as double moving averages, MACD golden crosses, and valuation ranking. Now, more users are seeing concepts closer to the institutional context for the first time, including IC, IR, layered backtesting, portfolio optimization, VaR, CVaR, Monte Carlo, Risk Parity and Black-Litterman.

One practitioner commented that WorkBuddy's quantitative Skills matrix repackages the "factor research + validity test + portfolio backtesting" scientific research workflow that originally only the quantitative research teams of securities firms and quantitative private fund research departments had, into capabilities that ordinary users can drive with natural language.

Five traditional thresholds of quantitative investment are removed

According to the reporter's communication with industry insiders, quantitative investment has long been "unattainable" for retail investors, and the crux lies in five thresholds:

The first is the data threshold, where the annual fee for professional terminals often ranges from tens of thousands to hundreds of thousands of yuan;

The second is the programming threshold, where users must master Python and quantitative frameworks;

The third is the threshold of professional knowledge and methodology, such as factor testing and overfitting identification;

The fourth is the tool cost threshold of purchasing paid platforms or renting servers;

The fifth is the cognitive discipline threshold brought by emotion and cognitive bias, as retail investors are more likely to chase rising stocks and sell off falling ones.

The emergence of quantitative investment-related Skills shows a trend of breaking these five thresholds one by one. Data can be accessed at any time through connectors, no programming technology is required as codes are generated by the Agent, methodologies are encapsulated and internalized in the Skills, backtesting is built-in and conclusions can be cross-verified, and at the cognitive level, the rule of "not relying on feelings and executing according to rules" is solidified into reproducible rules.

In a word, the barrier of quantitative investment for retail investors has never been "lacking a buy and sell signal", but "lacking a set of research infrastructure that is accessible, understandable and verifiable". These Skills have reduced the construction cost of such infrastructure from "hundreds of thousands of yuan + several months" to "one sentence".

Apart from the prosperity of the Skills ecosystem, what does the industry think of it? Multiple quantitative private fund investment managers and practitioners interviewed by CLS reporters gave calm judgments from different perspectives.

The tool threshold is removed first, and the research threshold is also declining

One point widely recognized by industry insiders is that AI Agent and quantitative Skills have indeed changed the way individual investors access quantitative investment.

A quantitative private fund investment manager told reporters that the popularization of Agent and quantitative Skills essentially reduces the threshold of quantitative investment "tools", rather than reducing the difficulty of investment itself. In the past, quantitative practitioners relied on programming capabilities and professional terminals to conduct data query, factor screening, backtesting framework construction and other work, but now they can complete these tasks with the help of quantitative Skills launched on the Agent.

This person said that for individual investors, quantitative Skills play a positive role in improving information acquisition efficiency, assisting data sorting, quickly screening targets and building a more structured research framework, but they cannot be directly used to improve investment returns.

The core difficulties of quantitative investment lie in data processing, factor mining, model strategy iteration, transaction execution, risk control and other links, which require long-term methodology accumulation, disciplined execution, and technical precipitation, and these are parts that Agent is difficult to replace at present.

Some fund managers also mentioned that encapsulating complex investment research processes and strategy logic into standardized modules can undertake a large number of repetitive tasks, and also reduce the interference of emotional trading to a certain extent.

What exactly does quantitative equalization equalize?

This is where the controversy arises. The fact that quantitative Skills make it easier for users to access quantitative investment does not mean that individual investors will have institutional-level capabilities from now on.

Many interviewees emphasized that the more accurate statement at present is equalization of knowledge and tools, rather than equalization of returns. One practitioner put it bluntly: these Skills provide retail investors with professional Beta and Smart Beta, not Alpha. They can help investors trade less by feeling, and establish rules and disciplines, but there is still a long way to go to stably outperform the market.

The reason is not complicated. The source of Alpha is often the information advantage, model advantage, execution advantage or trading structure advantage that others do not have. Once a strategy is written into a public Skill and called by more and more users, it will be difficult to maintain its scarcity. Public tools can spread methodologies and disciplines, but it is difficult to copy the most scarce excess returns along with them.

Institutional advantages go far beyond tools

This is also the boundary repeatedly reminded by industry insiders. Institutional quantitative investors have a complete link of factors, optimization, risk control and trading. Before strategies are launched, they need to go through noise filtering, overfitting identification, out-of-sample testing and capacity evaluation. "At this stage, even for quantitative practitioners, Agent is more like a work assistant that improves investment research efficiency," an investment manager summarized.

In addition, the advantages of institutional investors are far more than the tools themselves, but also lie in the long-term accumulated data system, computing power infrastructure, investment research framework and risk control discipline.

This gap will eventually fall into hard links, including data, execution, risk control and backtesting authenticity. Institutions can use tick-level data, order-by-order data, Level-2 data and more alternative data, and also have higher computing power, more mature research frameworks and stricter risk control disciplines. At the execution level, capabilities such as order placement through files, low-latency trading and arbitrage are still areas that retail investors cannot easily access through Agent and Skills.

But this does not weaken the value of quantitative Skills.

Quantitative practitioners said that platforms such as WorkBuddy encapsulate the methodology of quantitative research into tools that individuals can use directly, so that methods that were originally mastered more by institutions have truly entered the daily life of ordinary people for the first time. "In the final analysis, what quantitative investment relies on is nothing more than a set of clear and traceable rules and methods; and what this type of Skill does is to deliver these rules and methods to more people in the form of tools."

For most ordinary investors, they were not even familiar with concepts such as maximum drawdown, Sharpe ratio, factor orthogonalization and stress testing in the past. A tool that helps users understand IC, layered backtesting and stress testing is itself part of investor education.

One quantitative practitioner put forward a more pragmatic judgment: the equalization truly promoted by Skills targets the information disadvantage and cognitive disadvantage of retail investors relative to their past selves, rather than the capability gap between retail investors and institutions. Its end point is not necessarily to let individual investors have the same machine as institutions, and its more realistic value is to let ordinary people examine every transaction they make with a thinking framework closer to that of institutions.

This sentence may be closer to the reality at this stage. Quantitative Skills allow users to place bets less by emotion, verify ideas faster, and make trading disciplines clearer. They can help investors lose less to themselves. As for outperforming professional quantitative institutions, that remains another matter.

The future will move towards verticalization and will be more restricted by compliance

From the perspective of product evolution, the next step of quantitative Skills will most likely continue to deepen in several directions.

Data upgrading, professionalization of backtesting, portfolio-level risk control and scenario verticalization will all become subsequent directions. Minute-level data, Level-2 data and order-by-order data can correct the current roughness; transaction cost, impact cost, slippage and capacity models can reduce backtesting illusions; position health check, portfolio stress testing and intraday early warning are closer to real risk control needs.

A more sensitive step is transaction execution. Some practitioners judge that quantitative Skills may evolve from signal tools to semi-automatic execution in the future, accessing broker APIs to connect the whole chain from strategy to order placement. However, under the domestic regulatory environment, there are clear red lines for AI stock recommendation, client asset management on behalf of others, and programmatic transaction reporting. In the short term, the more realistic form is still to provide signals and explanations, and then let investors confirm the transactions.

Therefore, as quantitative Skills develop further, two questions need to be answered. One is the professional issue: whether the output can withstand the inspection of data caliber, backtesting assumptions and real trading constraints; the other is the compliance issue: whether the tool is a research aid, investment education, strategy signal, or will touch investment advice and automatic trading.

This article is from the WeChat official account "ChiNext Observer", author: Wang Chen, published with authorization from 36Kr.