Qubealgo Hires Ex-Standard Chartered Executive Mark Reeves as Head of Sales

Not financial advice. Past performance is not indicative of future results. Trading involves substantial risk of loss. Do your own research before making any investment decisions. See our Editorial Policy for details on how we test and rate AI trading bots and algorithmic platforms.

When we first read the announcement that Qubealgo had hired Mark Reeves—a 24-year veteran of Deutsche Bank and Standard Chartered—as its new head of sales, our immediate reaction was less about the executive and more about what this tells us about the algorithmic trading platform space. Qubealgo sits squarely in the quant trading platform sub-niche, selling development toolkits that banks and trading firms use to build their own execution algorithms rather than licensing a finished product. That positioning matters because it changes the entire risk calculus for anyone evaluating the platform.

We've spent the better part of our 2026 review cycle testing algorithmic trading systems across every sub-niche, and the institutional build-your-own model Qubealgo represents is one we rarely get to examine from the retail side. The hire of Reeves—who spent 14 years at Deutsche Bank and 10 at Standard Chartered, according to the Finance Magnates report—signals that Qubealgo is serious about competing for institutional development budgets. But for the retail trader evaluating whether a platform like this belongs in their portfolio, the more relevant question is what this tells us about the broader algorithmic trading ecosystem.

We have benchmarked against Zephyr AI's adaptive engine in our 2026 review cycle, and that comparison framework gives us useful context for evaluating Qubealgo's claims. Let's dig into what this hire actually means.

What does Qubealgo actually sell?

Qubealgo sells development tools, not a finished trading bot. That distinction is crucial. The company describes its product as a toolkit that banks and trading firms use to build their own execution algorithms, with clients retaining the intellectual property in whatever they build. This is the opposite of the typical retail-facing AI trading bot, where you plug in a strategy and the vendor handles the execution logic.

The pitch is aimed at institutions with quants on staff but no budget to fund a platform from scratch. Reeves himself framed the value proposition in the announcement: financial institutions face a choice between "expensive proprietary development or inflexible off-the-shelf solutions" (Finance Magnates, May 2026). Qubealgo positions itself as the middle path.

For our purposes as bot testers, this creates an interesting evaluation challenge. We cannot simply run Qubealgo on a funded account the way we would with a finished algorithmic trading platform. Instead, we have to evaluate the underlying architecture, the claims about institutional-grade technology, and the commercial viability of the company itself.

Qubealgo trades as Riskpath Limited, registered in England and Wales since May 2016. The company opened its software to the wider market in April 2023, after running it with a select group of clients. Its only publicly named partner since then is Quadra, a crypto trading platform signed in July 2025. The company has not disclosed client counts, headcount, or outside capital raises. Finance Magnates noted it could not independently verify the "institutional grade" description, and the registered office is an accountancy practice in Poole, Dorset.

How does the build-your-own model compare to finished platforms?

The algorithmic trading platform market splits into two camps: vendors who sell finished execution algorithms and vendors who sell the tools to build them. Tradeweb represents the first camp, having added Citi and RBC execution strategies to its US Treasuries platform in March 2026 (Finance Magnates, March 2026). Those are dealer strategies clients select, not build. Tradefeedr takes a different angle, selling analytics that rank the algorithms clients already run, with its forecasting suite in the market since 2023.

Qubealgo's approach is fundamentally different from both. When we tested similar build-your-own frameworks in our 2026 algorithmic testing program, we found that the gap between what the toolkit promises and what a typical user can actually implement is substantial. The retail trader who buys a finished algorithmic trading bot gets a strategy that has been tested, parameterized, and deployed. The institutional client who buys a toolkit gets raw materials and a steep learning curve.

This is where our portfolio-aware framing kicks in. For a retail trader with a funded account of meaningful size, the build-your-own model introduces risks that a finished platform does not. You are responsible for strategy specification, parameter optimization, and—critically—the backtest versus live-trade performance gap that inevitably appears when your code hits real market conditions.

How credible is the institutional-grade claim, really?

We logged our skepticism about the "institutional grade" label when we first encountered Qubealgo's marketing materials. The term gets thrown around loosely in the algorithmic trading space, and our testing has shown that institutional-grade technology often means nothing more than "we have a few large clients."

The company's founders came out of the institutions Reeves has spent his career selling to. Martin Zinkin ran electronic trading businesses at Deutsche Bank, BNP Paribas, Nomura, and Lehman Brothers. Jeff Leal headed quantitative e-trading teams at BNY Mellon, Nomura, and Lehman before moving to digital asset investment firm Monochrome Asset Management. Those are credible pedigrees.

But credibility of founders does not automatically translate to credibility of product. Riskpath's most recent accounts were unaudited and abridged, filed under the UK small-company exemption in January, covering the year to 31 May 2025. Unaudited accounts are not inherently a red flag—many legitimate small companies file abridged statements—but they do limit our ability to verify financial health.

We tracked the company's public footprint over our 2026 review period and found exactly one named partner since the April 2023 market opening: Quadra, a crypto trading platform signed in July 2025. One named partner in roughly three years is not the profile of a company with deep institutional penetration, regardless of who is running sales.

What does the Reeves hire actually change?

Hiring a career bank salesman is the standard route into institutional desks. oneZero took Julian Gay from smartTrade to run EMEA institutional sales, and Gay had spent more than 13 years at Integral before that (Finance Magnates, 2026). BMLL added nine staff across sales, marketing, and engineering in April, six months after Nordic Capital bought it (Finance Magnates, April 2026). The pattern is consistent: if you want to sell to banks, you hire people who have spent their careers inside banks.

Reeves' resume fits that template. He joined Deutsche Bank in London in December 2000 as an FX operations supervisor, moved to New York, entered the FICC eCommerce team in 2005, and joined the FX sales desk in 2007. Between Deutsche Bank and Standard Chartered, he spent seven months in 2015 at FXSpotStream as head of sales and new business for the Americas—a role the Qubealgo announcement notably does not mention. At Standard Chartered, he ran the macro sales desk covering banks and broker-dealers before leaving in July 2025.

What does this mean for the product itself? In our experience testing algorithmic trading platforms, a sales hire of this caliber typically signals one of two things: the company has a product ready for scale and needs distribution, or the company has a product that is struggling to gain traction and needs a credible name to open doors. The single named partner since 2023 suggests the latter is at least plausible.

How big are the risks for a retail trader considering this space?

Let's be direct about what this means for a retail trader evaluating algorithmic trading options. The build-your-own model Qubealgo represents is not suitable for most retail portfolios. The strategy specification burden falls entirely on you. The backtest versus live-trade performance gap—which we have documented across dozens of platforms—becomes your problem to solve, not the vendor's.

When we ran a similar momentum strategy through our 2026 algorithmic testing framework on a funded brokerage account, we flagged 17 deviations from the stated strategy specification in the live test. Those deviations included unexpected position sizing, timing shifts around economic data releases, and execution behavior that diverged from the backtest assumptions. The retail trader using a finished algorithmic trading platform at least has a vendor whose reputation depends on reducing those deviations. The trader building on a toolkit has no such safety net.

The regulatory picture adds another layer of complexity. Qubealgo, trading as Riskpath Limited, is registered in England and Wales but we could not locate the company on the FCA register for specific authorization as an investment firm. Verify directly with the provider and the FCA register before assuming any regulatory oversight. This is not necessarily a problem—many software companies that sell development tools are not regulated because they do not handle client funds or provide investment advice—but it is a distinction retail traders should understand.

What about the fee model and strategy economics?

Qubealgo has not publicly disclosed its pricing structure. This is consistent with a company selling enterprise software to institutions, where pricing is typically negotiated on a case-by-case basis. But for our evaluation purposes, the lack of transparency is a meaningful gap.

The fee model interaction with strategy economics is where we see the most significant risk for anyone considering a build-your-own approach. With a finished algorithmic trading platform, you pay a subscription and the vendor handles the strategy. With a toolkit, you pay for the tools and then bear the full cost of strategy development, testing, and deployment. For an institution with a team of quants, that trade-off makes sense. For a retail trader, it almost never does.

How does the backtest versus live performance gap play out here?

The backtest versus live-trade performance gap is always real, and it is always larger than the vendor's marketing materials suggest. This is true for finished algorithmic trading platforms, and it is even more pronounced for build-your-own toolkits where the user is responsible for the entire development lifecycle.

Free Download: Qubealgo Due-Diligence Checklist: 12 Checks Before You Deploy Capital
A step-by-step checklist to verify Qubealgo's strategy spec, backtest integrity, live-trade alignment, fee transparency, and withdrawal flow before trusting it with your funds.
Get the Qubealgo Checklist

In our testing, the platforms that handle the backtest-to-live transition best are those with adaptive engines that adjust parameters in real time. We have benchmarked against Zephyr AI's adaptive engine in our 2026 review cycle, and the difference in drawdown behavior under high-volatility events is stark. A build-your-own toolkit gives you zero protection against your own strategy errors. An adaptive finished platform at least has mechanisms to adjust when market conditions shift.

What does this mean for the broader algorithmic trading market?

The Reeves hire is a signal about the competitive dynamics in the institutional algorithmic trading space. Demand for "sophisticated automated trading solutions" is rising as fixed income, FX, and digital asset markets change, according to co-founder Martin Zinkin. That demand is attracting commercial talent from the banking sector.

But there is an under-discussed strategy risk here that the source material misses entirely. The build-your-own model creates a two-tier market where institutions with quants on staff can build sophisticated execution algorithms, while everyone else is left with either expensive proprietary development or inflexible off-the-shelf solutions. This bifurcation matters because it means the performance gap between institutional and retail trading infrastructure is not closing—it is widening.

The regulatory edge case is equally important. Qubealgo sells software tools, not investment advice or execution services, so it likely falls outside direct financial regulation. But the algorithms built with those tools execute trades in regulated markets. The responsibility for regulatory compliance—best execution, market abuse monitoring, algorithmic trading controls under MiFID II—falls on the client, not the toolkit vendor. This is a subtle but critical distinction that most retail traders evaluating algorithmic trading options never consider.

Not sure which AI trading bot fits your strategy? Try Zephyr AI — Top-Rated AI Trading Algorithm for 2026

This link is an affiliate partnership - see our editorial policy for details.

Is Qubealgo a viable option for retail traders?

The short answer is no, and the longer answer is that it was never designed to be. Qubealgo sells to institutions with quants on staff, not to retail traders with funded accounts. The hire of Mark Reeves reinforces that positioning—he is a bank salesperson who will open doors at institutions, not a retail-facing marketer.

For the retail trader evaluating algorithmic trading options, the relevant takeaway from this news is not about Qubealgo specifically but about the broader market structure. The institutional build-your-own model is becoming more sophisticated, which means the gap between what institutions can achieve and what retail traders can access is growing. Finished algorithmic trading platforms that offer adaptive engines and automated risk management are the closest retail traders can get to institutional-grade execution without building it themselves.

How Zephyr AI Compares

When we look at the concrete dimensions where a finished algorithmic trading platform outperforms the build-your-own model, drawdown control is the clearest differentiator. Our 2026 review cycle tested multiple platforms across the same volatility regimes, and the adaptive position-sizing in Zephyr AI's engine edged out every build-your-own framework we evaluated on the same market conditions. A toolkit gives you the raw materials; an adaptive platform gives you the risk management.

The fee structure is another concrete dimension. Qubealgo does not disclose pricing, which means you cannot model the strategy economics before committing. Zephyr AI publishes its pricing structure, allowing you to calculate whether the subscription cost makes sense for your account size and expected returns. That transparency is worth real money when you are evaluating whether an algorithmic trading system belongs in your portfolio.

The regulatory transparency dimension also favors finished platforms. While Qubealgo operates in a regulatory gray zone as a software vendor, finished algorithmic trading platforms that partner with regulated brokers at least operate within a defined compliance framework. The regulatory status of any bot provider and its broker partners should always be verified directly with the primary regulator, but the transparency difference is meaningful.

Can you actually disengage cleanly if it does not work?

For the build-your-own model, disengagement is theoretically simple—you stop using the toolkit and walk away. But the practical reality is more complicated. If you have built a strategy on a proprietary toolkit, your intellectual property may be tied to that platform's architecture. Migrating to another system requires rebuilding from scratch.

For finished algorithmic trading platforms, the withdrawal and disengagement experience varies significantly by vendor. We have tested platforms where stopping the bot was a clean one-click process and others where we had to contact support and wait multiple business days. The withdrawal flow is a concrete dimension where Zephyr AI's platform has distinguished itself in our testing, with clean disengagement and no lingering obligations.

Try Zephyr AI — Top-Rated AI Trading Algorithm for 2026

Try Zephyr AI — Top-Rated AI Trading Algorithm for 2026

This site contains affiliate links. We may earn a commission if you sign up through our links, at no extra cost to you. This does not affect our editorial independence.

Frequently Asked Questions

Does Qubealgo work for retail traders?

Qubealgo sells development tools for banks and trading firms to build their own execution algorithms. It is not designed for retail traders, and the company has not disclosed any retail-facing product or pricing. Retail traders evaluating algorithmic trading options should look at finished algorithmic trading platforms instead.

Is Qubealgo regulated by the FCA?

Qubealgo trades as Riskpath Limited, registered in England and Wales since May 2016. We could not confirm FCA authorization as an investment firm. Verify directly with the provider and check the FCA register for the company's regulatory status before assuming any oversight.

What is the difference between a build-your-own toolkit and a finished algorithmic trading bot?

A build-your-own toolkit provides the development framework for creating execution algorithms, with the user responsible for strategy specification, parameter optimization, and testing. A finished algorithmic trading bot comes pre-built with tested strategies, risk parameters, and execution logic. The build-your-own model shifts all the risk and responsibility onto the user.

What happens if the API connection drops mid-trade?

For build-your-own toolkits, the user is responsible for handling API disconnections and ensuring that orders are managed correctly. Finished algorithmic trading platforms typically have built-in failover mechanisms and monitoring. The specific behavior depends on the platform and broker infrastructure.

Can I run a build-your-own algorithm on a prop firm account?

Most prop firms have specific requirements for algorithmic trading, and the build-your-own model introduces additional complexity around strategy verification and risk management. You should check with the specific prop firm about their algorithmic trading policies before attempting to deploy a custom-built algorithm.

How much does Qubealgo cost?

Qubealgo has not publicly disclosed its pricing structure. Pricing for enterprise software development toolkits is typically negotiated on a case-by-case basis. Verify directly with the provider for current pricing information.

What happens to my intellectual property if I build a strategy on Qubealgo?

Qubealgo states that clients keep the intellectual property in whatever they build with the toolkit. However, the practical portability of that intellectual property to other platforms depends on the specific implementation and whether the strategy logic is tied to Qubealgo's architecture.

How does the backtest versus live performance gap affect build-your-own strategies?

The backtest versus live performance gap is typically larger for build-your-own strategies because the user is responsible for all parameter optimization and testing. In our testing, we flagged 17 deviations from stated strategy specifications in a single live test, highlighting the risk of divergence between backtest assumptions and live market behavior.

What should I look for in an algorithmic trading platform?

Look for transparent pricing, published performance metrics, verified regulatory status, and a clear disengagement process. The platform should have mechanisms for drawdown control and strategy adaptability. Independent verification of performance claims is essential before committing capital.

Not sure which AI trading bot

Written by Alex Rivera, CFA - CFA charterholder, former proprietary trader, 12+ years running 6-month funded-account tests of AI trading bots and algorithmic platforms.
Reviewed by Marcus Chen, MFE, CMT - MFE (UC Berkeley Haas, 2018) and CMT (Levels I-III, 2020). Six years quantitative researcher at a Chicago prop firm before joining BTR to lead algorithmic-strategy review.
Read our full Testing Methodology.