The Blushing Quants Podcast

Roman Isachenko is a quantitative researcher with a background in applied mathematics and rocket science who moved from engineering into finance, derivatives, and systematic trading. His experience spans risk management, derivative pricing, asset management, and small hedge fund environments, giving him a grounded view of how quant research actually works when capital, time, and market reality put every idea under pressure.

In this episode, we talk about the realities of building and rebuilding quant strategies in an environment where alpha decays quickly, and competition keeps getting tougher. Roman shares his path from engineering into quant finance, explains why derivatives first pulled him into the field, and reflects on the difference between elegant theory and what survives in live markets. We discuss pricing models, calibration, volatility, and why even strong ideas can stop working when market structure changes.

We also get into research discipline, overfitting, noise testing, Monte Carlo comparisons, strategy decay, and the hard decisions small teams face when a model starts to fail. Roman breaks down the differences between research at large institutions and at small funds, why execution and exits often matter more than entry signals, and why leadership, judgment, and genuine passion still matter as much as technical skill in quantitative finance.

*DISCLAIMER*

The information shared on this podcast is for educational and informational purposes only and reflects the personal opinions of the hosts and guests at the time of recording. Nothing in this podcast constitutes financial, investment, legal, tax, or trading advice, and nothing should be interpreted as a recommendation to buy, sell, or hold any security, cryptocurrency, derivative, or financial product.

Trading and investing involve substantial risk, including the possible loss of all or part of your capital. You are solely responsible for your own decisions, and you should consult a qualified professional before making financial decisions. By listening to this podcast, you agree that the hosts, guests, and producers are not liable for any losses or damages arising from the use of any information discussed.

Zach Marx is the Chief Investment Officer of Vineyard Quant Capital, where he works at the intersection of systematic equity investing, institutional flow, and data-driven portfolio construction.

In this episode, we get into what it actually takes to build a quantitative investment process around how institutions and retail investors make decisions, and how that can be turned into a systematic equity strategy.

We talk about Zach’s path from market data and S&P to running an investment strategy at Vineyard, why understanding how different allocators behave can be just as important as understanding the data itself, and how his team uses point-in-time information, forward expectations, and cross-sectional ranking to scale a fundamentally informed process across thousands of stocks. We also get into sector- and industry-level clustering, orthogonal factor construction, seasonality, in-sample versus out-of-sample testing, stock-level and portfolio-level risk management, and why running a successful quant strategy is as much about operations, relationships, and business building as it is about research. The conversation also touches on retail sentiment, alpha-capture frameworks, and how Zach approaches using AI to evaluate processes, without yet trusting it to make investment decisions on its own.

*DISCLAIMER*

The information shared on this podcast is for educational and informational purposes only and reflects the personal opinions of the hosts and guests at the time of recording. Nothing in this podcast constitutes financial, investment, legal, tax, or trading advice, and nothing should be interpreted as a recommendation to buy, sell, or hold any security, cryptocurrency, derivative, or financial product.

Trading and investing involve substantial risk, including the possible loss of all or part of your capital. You are solely responsible for your own decisions, and you should consult a qualified professional before making financial decisions. By listening to this podcast, you agree that the hosts, guests, and producers are not liable for any losses or damages arising from the use of any information discussed.

Mark Aron Szulyovszky is a crypto quant and an entrepreneur focused on cross-sectional alpha factors in digital assets. He works to surface crypto-native factors, build market-neutral portfolios, and turn research on derivatives, microstructure, and token-specific behavior into tradable products for both internal use and external clients.

In this episode, we get into what it actually takes to build and trade crypto-native alpha factors in a market that is volatile, fragmented, and still structurally different from traditional finance.

We talk about why Mark moved away from a more machine-learning-heavy approach toward simpler, more interpretable factor research, how he thinks about market microstructure and derivatives data in crypto, and why cross-sectional alpha remains more abundant there than in many traditional markets. He explains how his team classifies and cleans data, works with tradable universes, controls turnover, and looks for signals that can survive transaction costs rather than just look good in backtests. We also get into retail versus institutional flow, the relationship between spot and perpetual futures, portfolio construction under extreme non-stationarity, and why, in crypto, the biggest challenge is often not finding a signal but sizing and risk-managing it properly when the market regime shifts. The conversation also touches on his entrepreneurial path, including selling his first company and applying that builder mindset to quantitative finance.

*DISCLAIMER*

The information shared on this podcast is for educational and informational purposes only and reflects the personal opinions of the hosts and guests at the time of recording. Nothing in this podcast constitutes financial, investment, legal, tax, or trading advice, and nothing should be interpreted as a recommendation to buy, sell, or hold any security, cryptocurrency, derivative, or financial product.

Trading and investing involve substantial risk, including the possible loss of all or part of your capital. You are solely responsible for your own decisions, and you should consult a qualified professional before making financial decisions. By listening to this podcast, you agree that the hosts, guests, and producers are not liable for any losses or damages arising from the use of any information discussed.

Manuel Ritsch is the founder of Alpha Rho Technologies, where he is building AI-native investment infrastructure for asset management. After seeing how much of the industry still relied on outdated tools and manual processes, he set out to replicate the work of human analysts with AI and turn that into a real operating model for funds.

In this episode, we step slightly outside pure quant research to explore what it actually takes to build an AI-driven investment business from the ground up.

We talk about the gap between traditional asset managers and newer AI-native approaches, why low-frequency investing may be one of the clearest use cases for agentic systems, and how Manuel structured a fund run by AI analysts, CIOs, and investment committees. We also get into go-to-market, fundraising, bank partnerships, product positioning, client education, and the hard reality that institutional investors still want a track record, transparency, and a process they can trust. Manuel also explains why explainability matters so much in asset management, why models themselves are becoming commodities, and why the real edge increasingly comes from orchestration, usage, and product design rather than just model access.

*DISCLAIMER*

The information shared on this podcast is for educational and informational purposes only and reflects the personal opinions of the hosts and guests at the time of recording. Nothing in this podcast constitutes financial, investment, legal, tax, or trading advice, and nothing should be interpreted as a recommendation to buy, sell, or hold any security, cryptocurrency, derivative, or financial product.

Trading and investing involve substantial risk, including the possible loss of all or part of your capital. You are solely responsible for your own decisions, and you should consult a qualified professional before making financial decisions. By listening to this podcast, you agree that the hosts, guests, and producers are not liable for any losses or damages arising from the use of any information discussed.

Francisco Prack is a quant, economist, and portfolio manager with 30+ years of experience across financial markets, and a background spanning traditional finance, quantitative research, algorithmic trading, and crypto.

In this episode, we get into how a deeply model-driven way of thinking can shape an entire career in markets, from economics and traditional finance to algorithmic trading, reinforcement learning, and crypto.

We talk about why Francisco sees markets as a sequential problem rather than a static one, how he studies the tape day by day to extract patterns, and why understanding market rules and order types matters before touching the data at all. He explains how he thinks about institutional footprints, why replaying and re-reading past market sequences can be more useful than forcing generic statistical frameworks onto trading, and how reinforcement learning fits into his process by helping adapt parameter choices and position sizing across different market conditions. We also get into the practical differences between TradFi and crypto, the importance of writing conservative code for extreme market events, and why he still prefers to write the core logic himself rather than outsource the brain of the system.

*DISCLAIMER*

The information shared on this podcast is for educational and informational purposes only and reflects the personal opinions of the hosts and guests at the time of recording. Nothing in this podcast constitutes financial, investment, legal, tax, or trading advice, and nothing should be interpreted as a recommendation to buy, sell, or hold any security, cryptocurrency, derivative, or financial product.

Trading and investing involve substantial risk, including the possible loss of all or part of your capital. You are solely responsible for your own decisions, and you should consult a qualified professional before making financial decisions. By listening to this podcast, you agree that the hosts, guests, and producers are not liable for any losses or damages arising from the use of any information discussed.

Denis Lukyanov is a quantitative researcher and AI/ML practitioner working at the intersection of finance, machine learning, and agentic systems.

In this episode, we get into what it really takes to integrate agentic systems and large language models into quant workflows, and why the hard part is not generating ideas quickly, but building something structured, testable, and useful in practice.

We talk about the gap between quick AI prototypes and production-grade systems, why planning still matters more than coding speed, and how domain expertise remains the real bottleneck even as the tools improve. Denis breaks down how he thinks about combining traditional machine learning and deep learning with GenAI agents: LLMs can add real value as orchestrators, analysts, and research accelerators, but they still should not be trusted to make decisions. We also get into context windows, knowledge systems, guardrails, model-as-judge workflows, regime detection, quantitative research loops, and why serious trading systems still need explicit logic, strong data, and human control.

If you care about how AI is actually being used inside quant research today, what separates real systems from AI theater, and how to think clearly about agents, models, and market structure without losing rigor, don’t miss this one.

*DISCLAIMER*

The information shared on this podcast is for educational and informational purposes only and reflects the personal opinions of the hosts and guests at the time of recording. Nothing in this podcast constitutes financial, investment, legal, tax, or trading advice, and nothing should be interpreted as a recommendation to buy, sell, or hold any security, cryptocurrency, derivative, or financial product.

Trading and investing involve substantial risk, including the possible loss of all or part of your capital. You are solely responsible for your own decisions, and you should consult a qualified professional before making financial decisions. By listening to this podcast, you agree that the hosts, guests, and producers are not liable for any losses or damages arising from the use of any information discussed.

Toby Morris works across multi-asset trading, client coverage, sales trading, and trading desk operations, helping clients execute effectively while keeping the desk, workflow, and decision-making process aligned behind the scenes.

In this episode, we go beyond job titles to explore what the trading desk actually looks like when clients, liquidity, technology, and judgment collide in real time.

We talk about how markets evolved from phone-brokered flow to online and mobile trading, and what that shift changed for both clients and trading desks. Toby shares a grounded view on what it really takes to deliver for clients in practice: understanding their needs, choosing the right execution approach, knowing when to systematize, and knowing when discretion creates more risk than value. We also get into the hidden operational side of the business, from aligning teams across trading, operations, risk, and compliance to handling large orders, internal communication, liquidity constraints, and the uncomfortable reality that sometimes the most important skill is knowing when to say no.

*DISCLAIMER*

The information shared on this podcast is for educational and informational purposes only and reflects the personal opinions of the hosts and guests at the time of recording. Nothing in this podcast constitutes financial, investment, legal, tax, or trading advice, and nothing should be interpreted as a recommendation to buy, sell, or hold any security, cryptocurrency, derivative, or financial product.

Trading and investing involve substantial risk, including the possible loss of all or part of your capital. You are solely responsible for your own decisions, and you should consult a qualified professional before making financial decisions. By listening to this podcast, you agree that the hosts, guests, and producers are not liable for any losses or damages arising from the use of any information discussed.

Mattia Spreafico is a quant based in Switzerland with an MSc in Quant Finance and a background in Mathematical Engineering from Politecnico di Milano.

In this episode, we go beyond job titles and get into what the next generation of quants is actually dealing with day to day inside large institutions, where speed, correctness, and deployment constraints collide.

We talk about how AI is already changing the quant workflow in practice, not as hype but as a default tool for research, coding, and iteration. Mattia shares a grounded view on where LLMs genuinely help, where they still fall short in finance, and why data quality and access are still the real foundation. We also dig into the uncomfortable tradeoff nobody wants to admit: a perfect review slows you down, but moving fast increases risk, so the game becomes building better monitoring, better controls, and faster reaction when something breaks.

If you care about how quant research is evolving from idea to production, why deployment can be harder than modeling, and what skills will matter most when everyone can “code” but fewer can design robust systems, this conversation will hit.

*DISCLAIMER*

The information shared on this podcast is for educational and informational purposes only and reflects the personal opinions of the hosts and guests at the time of recording. Nothing in this podcast constitutes financial, investment, legal, tax, or trading advice, and nothing should be interpreted as a recommendation to buy, sell, or hold any security, cryptocurrency, derivative, or financial product.

Trading and investing involve substantial risk, including the possible loss of all or part of your capital. You are solely responsible for your own decisions, and you should consult a qualified professional before making financial decisions. By listening to this podcast, you agree that the hosts, guests, and producers are not liable for any losses or damages arising from the use of any information discussed.

Robert is a London-based systematic trader with 25+ years in markets. He started in the early 2000s prop trading futures, survived the no-simulator era, and evolved from discretionary trading into fully systematic research and automation. Today, he builds short-term strategies across equity indices and rates futures, and has recently helped a large institution stand up a proprietary trading team.

In this episode, we get practical about how an independent trader thinks and operates: finding edges in intermarket relationships, turning market intuition into systematic decision trees, and building portfolios that aim for strong Sharpe with positive skew. Rob also breaks down how he approaches regime awareness and robustness, in-sample vs out-of-sample work, and why walk-forward style processes are harder than they sound when you actually care about “the right trades", not just the best backtest. We also talk about the fundraising catch-22 for independents and why selling signals and research can be a smarter wedge than trying to raise a big flagship fund on day one.

*DISCLAIMER*

The information shared on this podcast is for educational and informational purposes only and reflects the personal opinions of the hosts and guests at the time of recording. Nothing in this podcast constitutes financial, investment, legal, tax, or trading advice, and nothing should be interpreted as a recommendation to buy, sell, or hold any security, cryptocurrency, derivative, or financial product.

Trading and investing involve substantial risk, including the possible loss of all or part of your capital. You are solely responsible for your own decisions, and you should consult a qualified professional before making financial decisions. By listening to this podcast, you agree that the hosts, guests, and producers are not liable for any losses or damages arising from the use of any information discussed.

Haris is a quantitative equity researcher and portfolio manager with experience across top-tier institutions, including BlackRock. Born in Greece, he studied applied computer science and applied mathematics, worked at the European Central Bank, then moved to the US for a Master's in Financial Engineering, and later built systematic equity models in the industry.

In this episode, we go into how quant research is done when time matters. Haris breaks down the prototype mindset, how to build a fast research pipeline, and what gets rejected immediately before you waste months. We talk about signal evaluation, residual momentum, sector and risk neutralization, and why correlation with your existing book can kill a "great" signal.

We also unpack execution realities, trading costs, backtest vs. realized PnL, Monte Carlo and holdouts, and the optimizer's approach to sizing, constraints, turnover, and robustness. Finally, we cover where machine learning and deep learning fit, why short horizons give data-hungry models an edge, and how LLM agents are accelerating the pace of research.

*DISCLAIMER*

The information shared on this podcast is for educational and informational purposes only and reflects the personal opinions of the hosts and guests at the time of recording. Nothing in this podcast constitutes financial, investment, legal, tax, or trading advice, and nothing should be interpreted as a recommendation to buy, sell, or hold any security, cryptocurrency, derivative, or financial product.

Trading and investing involve substantial risk, including the possible loss of all or part of your capital. You are solely responsible for your own decisions, and you should consult a qualified professional before making financial decisions. By listening to this podcast, you agree that the hosts, guests, and producers are not liable for any losses or damages arising from the use of any information discussed.