Come meet the YellowDog team and discover how quantitative research teams can accelerate Monte Carlo simulations and complex model workloads—scaling seamlessly from local environments to thousands of cloud nodes with zero infrastructure friction. Using the popular Ray framework, we’ll show how YellowDog’s intelligent orchestration and autoscaling deliver the speed, precision, and elasticity today’s quants need to achieve high-scale performance without complexity.

Demo Drive led by: Alan Parry, CTO, YellowDog

Booth: B5

Sponsored by: YellowDog

---

• Is AI’s impact a natural extension of existing quantitative methods, or a fundamental disruption? Explore AI’s capabilities through discussions on:
• What recent advancements in AI and machine learning have significantly improved alpha signal extraction, and how are these integrated into systematic strategies?
• How are reinforcement learning, predictive modeling, and execution algorithms driving measurable improvements in portfolio execution efficiency?
• What distinguishes incremental model enhancements from disruptive AI techniques in the context of strategy evolution and market adaptability?
• How are adaptive AI frameworks—such as agent-based models and meta-learning systems—shaping the future of autonomous trading in complex, dynamic environments?

Federico Fontana

Chief Technology Officer
XAI Asset Management

See Bio

Aditya Sharma

Head of Textual and Quantitative Signals
S&P Global Market Intelligence

See Bio

Revant Nayar

Principal and CIO
Princeton AI and FMI Tech

See Bio

Sid Pednekar

Research Analyst
Millennium

See Bio

A practical exploration of how leading firms navigate risk and complexity to enhance returns through quantitative investment strategies.

This session should attract quant/fund managers/allocators and large pension funds and insurance firms interested in understanding how major asset managers (FoF allocators) are leveraging quantitative investing within their portfolio strategies.

Aous Labbane

Founder
Jasmin Capital and Consultancy

See Bio

Mathias Piardon

CEO
1L Capital AG

See Bio

Piotr Podgorny

Director, Quantitative Research
Natixis Investment Managers

See Bio

Machine learning has become increasingly central to both forecasting and portfolio construction. But how do you separate hype from real, actionable value?

• What ML models are most effective for return and risk forecasting?
• How are firms integrating AI and ML signals into quant investment strategies?
• When does ML add value over traditional quant techniques?

Dr. Joo Hee Lee

Consultant
CFA Institute

See Bio

Hamza Bahaji

Head of Financial Engineering & Investment Solutions
Amundi

See Bio

Andrea Nardon

CIO
Creed & Bear

See Bio

Leon He

Managing Director, Quantitative Equity
BCI

See Bio

Traditional risk models often fall short in capturing the complex dynamics across multiple asset classes, regions, and liquidity profiles. Gain insights into:

• The development and implementation of functional, adaptable, data-driven risk models
• Cross-asset volatility, as well as fact and correlation modelling.
• Practical challenges in stress testing and scenario analysis.

Panel Questions:

• How can we develop and implement functional, adaptable, data-driven risk models that effectively capture the complexities across multiple asset classes, regions, and liquidity profiles?
• In the context of cross-asset volatility, what are the best practices for factor and correlation modeling to ensure accurate risk assessment and management?
• What are the practical challenges faced in conducting stress testing and scenario analysis, and how can these be addressed to improve risk model robustness?
• Given the dynamic nature of financial markets, how can risk models be designed to adapt to changing conditions and provide reliable insights during periods of market stress?

Dr. Joo Hee Lee

Consultant
CFA Institute

See Bio

Florian Lelpo

Head of Macro and Mult-Asset Portfolio Manager
Lombard Odier Investment Managers

See Bio

Akhil Khunger

VP Quantitative Analytics
BARCLAYS

See Bio

Christoph Schon

Head of Investment Decision Research, EMEA
SimCorp

See Bio

Mathematical portfolio optimization is a tool for maximizing the expected return, minimizing the risk, or optimizing related measures for a portfolio of investments. Quantitative analysts and portfolio managers use portfolio optimization software to support them in making investment decisions. With Gurobi’s MIP technology it is possible to incorporate discrete decisions in the portfolio selection, like fixed costs, transaction limits, and cardinality constraints. The effectiveness of backtesting with discrete constraints is heavily dependent on the solver performance: The more strategies and scenarios can be run, the more alpha. In this session we will share best practices for modeling, implementing and tuning discrete decision models for optimal performance.

Robert Luce

Principal Developer
Gurobi Optimization

See Bio

Machine learning has become increasingly central to both forecasting and portfolio construction. But how do you separate hype from real, actionable value?

• What ML models are most effective for return and risk forecasting?
• How are firms integrating AI and ML signals into quant investment strategies?
• When does ML add value over traditional quant techniques?

Revant Nayar

Principal and CIO
Princeton AI and FMI Tech

See Bio

Dara Sosulski

Managing Director, Head of Artificial Intelligence and Model Management, MSS, HS
HSBC

See Bio

Naman Jain

Senior Portfolio Manager
Millennium Capital Partners LLP

See Bio

This hands-on workshop guides quant analysts through the complete workflow of utilizing KDB and q, the high-performance time-series database and programming language widely adopted in financial institutions. Participants will learn the fundamentals of KDB and q for handling real-time market data at scale, build and test end-to-end workflows for analysis, signal generation, and execution, and gain practical experience in writing, optimizing, and deploying production-ready code in a quant environment.

Session Objectives:

• Understand the core concepts and syntax of KDB and q for efficient data handling.
• Develop and test complete workflows for quantitative analysis and trading strategies.
• Acquire practical skills in optimizing and deploying code suitable for production environments in finance.

Session Breakdown (90 Minutes):

1. Introduction to KDB and q (15 minutes)

• Overview of KDB and q: architecture, features, and advantages in financial data processing.
• Discussion on the role of KDB/q in high-frequency trading and real-time analytics.

2. Hands-On Coding: Data Ingestion and Processing (25 minutes)

• Demonstration of ingesting real-time market data into KDB.
• Writing q scripts for data parsing, cleaning, and transformation.
• Best practices for handling large-scale time-series data.

3. Workflow Development: Analysis to Execution (25 minutes)

• Building analytical models and signal generation using q.
• Integrating analysis with execution systems.
• Testing and validating workflows in a simulated environment.

4. Optimization and Deployment (20 minutes)

• Techniques for optimizing q code for performance and scalability.
• Deploying workflows into production environments.
• Monitoring and maintaining deployed applications.

5. Q&A and Discussion (5 minutes)

• Open floor for participant questions
• Discussion on advanced topics and real-world applications.

Key Takeaways:

• Proficiency in using KDB and q for real-time data processing and analysis.
• Ability to develop end-to-end workflows from data ingestion to execution.
• Knowledge of best practices for optimizing and deploying quant applications in production settings.

Alexander Unterrainer

KDB Enthusiast
DefconQ

See Bio

Quantitative research teams rely on massive datasets, but file-based lakehouse stacks (Parquet + Iceberg) impose systemic bottlenecks—slow ingest, high-latency queries, and painful schema evolution. VAST DataBase collapses these layers into a unified, flash-native system. In this demo, we’ll show how Sorted Tables, V-Trees, and cell-level updates deliver millisecond queries, 21× faster backfills, and exabyte scale. The result: CPUs and GPUs stay fed, researchers iterate faster, and the data layer becomes a competitive advantage.

Demo Drive led by: Andrius Markvaldas, Senior System Engineer, VAST Data

Booth: C5

Sponsored by: VAST Data

---

1. Time-Series Modeling of Order Flow Using Deep Learning:

Demonstrate how deep learning models, when applied to high-frequency, structured time-series data in kdb+/q, can enhance short-term price forecasting and execution decision-making.

2. Combining Domain Knowledge with Neural Architectures:

Discuss how deep learning complements traditional signal processing in kdb+/q, addressing interpretability, robustness, and latency trade-offs in real-world execution pipelines.

3. Real-Time Execution Analytics at Scale:

Present techniques for efficient venue benchmarking, slippage monitoring, and dynamic strategy refinement using kdb+/q, aligned with live market microstructure feedback.

These objectives aim to bridge cutting-edge AI applications with time-critical systems built on kdb+/q, offering practical value to quants, engineers, and traders alike.

Rohit Sharma

Vice President, Market Surveillance
BNY Mellon Investment Management

See Bio

Large Language Models and NLP’s have moved from labs to trading desks—shaping the way we trade:

1.    Can you give real-world applications of LLMs in quant finance—from sentiment analysis to automated document parsing?

2.    How are you using NLP to accelerate research, generate alpha, and improve operational efficiency?

3.    How have you grasped the strengths, limitations, and responsible deployment of LLMs in production environments?

Samuel Livingstone

Head of Quantitative Strategies and Risk
Ambienta

See Bio

Terri van der Zwan PhD

Quantitative Researcher
Robeco

See Bio

Guido Baltussen

Head of Quant
Northern Trust Asset Management

See Bio

Alex Remorov

Director, Systematic Active Equities
BlackRock

See Bio

Dr Yichuan Zhang

CEO and Chief Research Officer
Boltzbit

See Bio

·     Time series models are at the core of forecasting in quantitative finance—powering everything from trading signals to risk management strategies:

·     Explore key models and deep learning approaches for financial forecasting

·     Discover how time series forecasts help you make data-driven decisions on timing, allocation, and risk

Wafaa Schiefler

ED, Commodities Quantitative Researcher
JP Morgan Chase

See Bio

Go beyond the theory and deep into the real-world application of quant innovation with Matthias Uhl, Head of Analytics & Quant Modelling (AQM) at UBS Asset Management. This session unpacks practical use cases where AI, machine learning, and NLP are actively being used to extract alpha, generate alternative data insights, and forecast market movements with greater precision.

Drawing from UBS’s institutional experience, Matthias will reveal how cutting-edge quant tools are built, tested, and deployed at scale — and what it means for the future of data-driven investing.

Audience Takeaways:

·      Discover real examples of how alternative data and NLP are used to generate tradable signals

·      Understand the ML methods and quant analytics driving next-generation forecasting models

·      Learn how UBS translates innovative research into practical, scalable investment tools

Matthias Uhl

Head Analytics & Quant Solutions, Partnership Solutions
UBS Asset Management

See Bio

As algo wheels reshape how buy-side firms select brokers, top quality algo performance is no longer a competitive edge—it’s a baseline requirement. Generic factors like bid-ask spread, volatility, top of the book liquidity, order size, etc., are no longer sufficient to navigate today’s intricate market dynamics. This session introduces “Lenses,” a powerful, new framework designed for the dynamic evaluation and enhancement of algorithmic trading performance.

Lenses empower quantitative teams to move beyond traditional metrics, offering unprecedented granularity to "zoom in" on execution quality through tailored, actionable insights. We will provide real-world examples demonstrating how “Lenses” reveal critical opportunities to refine routing logic, determine optimal schedules, or instantaneous participation rates, optimize child order placement and tolerance bands, and intelligently adapt taking strategies in real-time. Discover practical methods for cleaning liquidity, shrewdly managing spread crossing, and effectively mitigating fading effects—all crucial for preserving and maximizing algo performance.

Attendees will learn how to leverage “Lenses” to achieve smarter, better execution quality, and secure a decisive competitive edge in the rapidly evolving landscape of algo wheels.

Overview Objectives:

Optimize Execution Strategy: Empower quant teams to fine-tune execution strategies by leveraging “Lenses” for superior algo performance.

Master Advanced Taking Logic: Implement advanced taking logic, including techniques for cleaning liquidity, intelligent spread crossing management, and effectively mitigating fading effects to safeguard and grow alpha.

Gain Broker Evaluation Advantage: Strategically position your algorithms for success within evolving broker evaluation frameworks, aligning with client-specific benchmarks through adaptive, data-driven performance insights.

Gabriel Tucci

Global Head of Equities Cash Quant Trading
Citi

See Bio

As the lines between public and private markets blur, the demands on portfolio construction models are rapidly evolving. In this keynote, Quant of the Year 2024, Dr. Artur Sepp, Global Head of Investment Services Quant Group at LGT Private Banking, challenges the conventional frameworks that underpin strategic and tactical asset allocation frameworks.

Drawing from pioneering work at the intersection of advanced mathematics and real-world portfolio engineering, Artur reveals how next-generation optimization techniques are enabling fully integrated, cross-asset portfolios—spanning bonds, equities, hedge funds, and private equity. This is not the old-school mean-variance paradigm—it's a reimagining of what quant can achieve in a world where liquidity, transparency, and return dynamics vary dramatically across asset classes.

Key takeaways:

·      A first-hand look at the quant frameworks powering multi-asset portfolios across public and private markets.

·      Insights into mathematical innovation beyond mean-variance—including clustering and sparse methods for estimation of covariance matrix, tactical signals, and adaptive optimization.

·      A roadmap for building future-fit investment architectures that respond to today’s structural shifts in markets and capital allocation.

Artur Sepp

Global Head of Investment Services Quant Group
LGT Bank

See Bio

This hands-on workshop delves into the theory and practical implementation of reinforcement learning (RL) in quantitative finance. Participants will gain insights into building reliable RL models for trading and portfolio management, explore real-world applications, and examine case studies from hedge funds and quant labs.

Session Breakdown (80 Minutes):

1. Introduction to Reinforcement Learning in Finance (10 minutes)

• Overview of RL concepts: agents, environments, states, actions, rewards.
• Differentiating RL from traditional machine learning approaches.
• Relevance of RL in trading, portfolio management, and market making.

2. Building Reliable RL Models for Trading (20 minutes)

• Selecting appropriate RL algorithms (e.g., Q-learning, Deep Q-Networks, Policy Gradients).
• Designing state and action spaces tailored to financial markets.
• Defining reward functions that align with trading objectives.
• Addressing challenges like overfitting, non-stationarity, and exploration-exploitation trade-offs.

3. Real-World Applications (20 minutes)

• Examination of RL implementations in hedge funds and quant labs.
• Strategies for mitigating risks in live trading systems.
• Lessons learned from deploying RL agents in real market conditions.

4. Applications in Asset Allocation, Trade Execution, and Market Making (15 minutes)

• Utilizing RL for dynamic asset allocation strategies.
• Enhancing trade execution through RL-driven decision-making.
• Implementing RL in market-making scenarios to optimize bid-ask spreads and inventory management.

5. Interactive Q&A and Discussion (15 minutes)

Open floor for participant questions.

• Discussion on emerging trends and future directions in RL for finance.
• Sharing resources for further learning and exploration.

Key Takeaways:

• A solid understanding of how reinforcement learning can be applied to various aspects of quantitative finance.
• Practical knowledge of building and deploying RL models for trading and portfolio management.
• Insights into real-world challenges and solutions from industry case studies.
• Awareness of the potential and limitations of RL in live trading environments.

Nicole Konigstein

Chief AI Officer, Head of AI & Quant Research
quantmate

See Bio

Samuel Fernández Lorenzo is a quantum physicist and PhD in quantum technologies, CEO and co-founder of Inspiration-Q, a fintech specialized in quantum-inspired computing. He was also an associate professor at IE University. During his career he has worked as a researcher in the financial sector, developing his work on applications of quantum computers and quantum algorithms for finance and artificial intelligence. He has published numerous scientific articles in international journals and is a recognized expert in the application of quantum technologies to the financial sector.

Demo Drive led by: Samuel Fernández, Founder and CEO, Inspiration Q

Booth: A2

Sponsored by: Inspiration Q

---

Where and how models are deployed can impact everything from execution speed to compliance.

In this session:

• Compare performance, scalability, and latency trade-offs between cloud and on-prem deployments
• How do the costs compare? How can you avoid lock-in?
• Does the increased use of Gen-AI/GPUs/XPUs change things?
• Understand regulatory, data security, and compliance considerations
• Can hybrid architectures give a best of both world’s solution for balancing costs, speed, control, and flexibility

Barry Fitzgerald

Co-Head of Front-office Engineering
Man Group

See Bio

Quant teams are under pressure to extract alpha from increasingly complex and unconventional datasets. As new signals, factors, and sources emerge, how do you build models that are not only predictive—but resilient and scalable?

·      How do we quantify data quality?

·      What are the impacts of data quality on quantitative models, e.g. can models work with dirty or missing data?

·      Learn how to integrate new data sources into model pipelines without compromising stability

·      Understand the evolving role of alternative data in sustainable alpha generation

·      Assessing signal trading and landing models, assessing processes and cleaning dirty data

·      What are the different use cases of data and data quality requests from various stakeholders, including business, validation, risk, and regulators?

Kristina Ūsaitė

Senior Quantitative Researcher
Robeco

See Bio

Giuliano De-Rossi

Head Quants Equities in the Chief Investment Office
UBS Global Wealth Management

See Bio

Erdem Ultanir

Credit Risk Manager
Barclays

See Bio

Maarten Smit

Quantitative Active Portfolio Analyst, Senior Vice President
Northern Trust Asset Management

See Bio

·      What are the most impactful advancements in multi-factor risk modelling, and how are they reshaping portfolio construction in high-volatility environments?

·      In what ways are AI, machine learning and quantum computing techniques enhancing or challenging traditional portfolio construction?

·      How can quants effectively integrate signal uncertainty and regime shifts into dynamic risk models and portfolio rebalancing strategies?

Carlos Arcila-Barrera

Portfolio Manager and Research Lead
Sigma Advanced Capital Management

See Bio

Giulio Renzi-Ricci

Head of asset allocation
Vanguard

See Bio

Alexis Yannakou

Quant Strategist
Citadel

See Bio

Samuel Livingstone

Head of Quantitative Strategies and Risk
Ambienta

See Bio

As digital assets mature beyond the hype cycle, a growing number of institutional investors are exploring systematic strategies to manage volatility and capture alpha in an inherently fragmented, fast-moving market.

Panel Questions:

1.    Can you elaborate on how you are developing the end-to-end architecture of a long/short quant crypto strategy—using fundamental and systematic analysis and CTA models and trading them systematically with quant models?

2.    In the context of a multi-strategy, institutional-grade digital asset fund that blends internal systematic strategies with external manager allocations, how do you approach portfolio construction and dynamic capital allocation across a diverse set of models and trading styles?

3.    What frameworks or technologies enable you to optimize risk-adjusted returns while managing cross-strategy correlation in a volatile and fragmented market?

Ankush Jain

Advisory
Jain Holdings

See Bio

Samed Bouaynaya

Portfolio Manager
Altana Wealth

See Bio

Jean-Marc Bonnefous

Managing partner
Tellurian Capital

See Bio

Alessandro Balata

Portfolio Manager
Fasanara Digital

See Bio

Evgeny Byvatov

Head of Systemic Volatility Trading
GSR

See Bio

As foundation models like GPT, Claude, and Gemini enter finance, the next quant frontier is extracting latent signals from massive unstructured datasets. This masterclass explores how LLMs can be fine-tuned and interrogated to extract market-moving insights from earnings calls, filings, news flow, and more. Participants will prototype alpha pipelines using LLM outputs and debate their interpretability, drift, and compliance viability.

Structure:

Part 1: Applied LLM Briefing

Walkthrough of practical use cases: fine-tuned models for sentiment extraction, risk signal detection, and meta-feature creation

Part 2: Group Alpha Sprint

Teams receive an unstructured dataset (e.g., snippets from 10-Ks, call transcripts, macro headlines) and a mission: generate 1–2 usable features or signals using a foundation model output (pre-provided or simulated)

Part 3: Debate Round

Teams justify their signal’s validity, interpretability, and potential drift exposure; the group votes on robustness

Part 4: Expert Close

Discussion on prompt engineering, retraining pipelines, and alpha lifecycle for LLM signals

Takeaways:

• Practical LLM integration in quant workflows
• Feature engineering from unstructured sources
• Risk assessment of opaque models in a high-compliance context

Sponsored by: Boltzbit

---

Dr Yichuan Zhang

CEO and Chief Research Officer
Boltzbit

See Bio