​Description

​The inaugural Tokyo AI (TAI) session on AI in Financial Services explores the intersection of high-performance computing, generative modeling, and production-grade machine learning within the financial sector. As the industry moves beyond traditional statistical methods, technical leaders face the dual challenge of leveraging increased computational power while mitigating model fragility and "black box" risks. This technical meetup features deep dives into the evolution of quantitative finance, the use of VAEs and LSTM-FLOW for synthetic market scenario generation, and the practical application of Shapley values for model explainability in production.

​​Agenda

​The sequence follows a "Macro to Micro" progression. Shota Ishii provides the essential historical and industry-wide context. Syed Hashim follows with specific, deep-tech generative architectures (VAEs/LSTMs) for risk. Andreas Alexelis concludes by addressing the critical requirement for any production model: interpretability and stakeholder trust.

• ​18:00: Doors open

• ​18:30 – 19:00: A Kaleidoscopic Survey of How AI is Transforming Finance (Shota Ishii)

• ​19:00 – 19:30: Enhancing Financial Risk Management with Generative AI and Machine Learning (Syed Hashim)

• ​19:30 – 20:00: Demystifying the Black Box: Practical Explainability in Financial Machine Learning (Andreas Alexelis)

• ​20:00 – 21:00: Networking

​Speakers

​Talk 1 - A Kaleidoscopic Survey of How AI is Transforming Finance

​Speakers: Shota Ishiii (CEO, ProssimoTech)

​Abstract: This talk surveys the evolution of quantitative finance through the lens of increasing computational power and the proliferation of data. It begins with a brief review of traditional approaches—such as statistical methods for covariance estimation—and traces how advances in artificial intelligence, high-performance computing, and alternative data are reshaping the discipline. We examine how these shifts are enabling new methodologies across core domains, including scenario generation and simulation, portfolio optimization, and research workflows. At the same time, the discussion addresses emerging risks: model over-complexity, operational fragility, and the growing incidence of low-quality or misapplied AI (“slop”) in financial contexts. Finally, the talk connects these technological changes to tangible business outcomes—ranging from alpha generation in asset management to advancements in risk management, corporate finance, and personal finance—offering a structured view of how the financial landscape is evolving in practice.

​Bio: Shota has over 25 years of experience at the intersection of finance, data science, and artificial intelligence. He focuses on the design and implementation of knowledge and decision systems for corporate finance and investment management.
He previously served as a Managing Director at State Street Global Exchange, where he helped build a San Francisco–based team applying AI and high-performance computing to next-generation portfolio analytics. Earlier in his career, he was a product specialist at DCI (now part of Blackstone Credit), focusing on corporate credit investing. His experience also includes founding a knowledge management startup in Paris and serving as Head of Asia for KMV LLC (now part of Moody’s Analytics), a pioneer in quantitative credit risk modeling. He has advised extensively on balance sheet optimization as part of the Financial Strategies Group at Shinsei Bank (now SBI Shinsei Bank). Shota holds a B.S. in Applied and Engineering Physics from Cornell University and an MBA from INSEAD. He is also a Fellow at Oliver Wyman.

​Talk 2 - Enhancing Financial Risk Management with Generative AI and Machine Learning

​Speaker: Syed Hashim (Senior Consultant, Aurora Solutions)

​Abstract: Financial risk management at institutions relies heavily on historical simulations. In this talk, we discuss the usage of machine learning and generative AI models—specifically VAEs and LSTM-FLOW—to probabilistically generate realistic, synthetic market scenarios while rigorously maintaining dynamic cross-asset dependencies. We show how these models can be used to validate existing in-house models, create novel stress scenarios, and identify specific market scenarios that might severely impact a given portfolio. We further discuss the use of anomaly detection techniques and ML-based approaches for solving and significantly speeding up non-linear and non-convex optimization problems, enhancing rates of finding global minima when fitting complex curves and volatility surfaces.

​Bio: Hashim is a Senior Consultant at Aurora Solutions K.K. in Tokyo, specializing in financial machine learning and generative modeling for risk management. Drawing on a multidisciplinary background in computational biophysics, statistical mechanics, and optimization, he applies advanced computational methodologies to solve complex financial risk management problems.

​Talk 3 - Demystifying the Black Box: Practical Explainability in Financial Machine Learning

​Speakers: Andreas Alexelis (CTO, AlpacaTech)

​Abstract: Predictive machine learning models are transforming portfolio allocation and exposure management, yet their "black box" nature often leaves investors hesitant. Even highly performant strategies require transparency to gain trust. This talk details how to successfully integrate explainability methodologies into the production monitoring phase of financial ML solutions. Highlighting real-world examples from our monitoring framework, we will explore how utilizing Shapley values and residual error analysis can decode complex model behaviors, turning opaque predictions into interpretable, trustworthy insights for stakeholders.

​Bio: Director CTO at AlpacaTech, overseeing the technical roadmap and solutions quality of the AI Business department. Carrier in directing multilingual, multicultural, cross-functional teams to achieve their technical and business objectives in FinTech, HR tech, and manufacturing industries, with a strong technical background and a hands-on management style in the context of a venture business environment. Served as VPoE at AlpacaJapan, previously in ExciteOne Inc. as information systems director in charge of FX platform integration, in Bizreach as the technical head of the service targeting the SE Asian market, and before that in Sumitomo Electric Inc. Co as assistant manager in the Overseas Group companies support department.

​​Organizers

​​​​​​​Ilya Kulyatin is an entrepreneur with work and academic experience in the US, Netherlands, Singapore, UK, and Japan. He holds a BA in Economics, an MA in Finance, and an MSc in Machine Learning. He's a 3x founder, now helping Japan grow the local AI ecosystem through a not-for-profit community, Tokyo AI (TAI), while building an AI-native system integrator and solutions provider, Foundry Labs株式会社.

​​Supporters

​​​​Tokyo AI (​​​TAI) is the biggest AI community in Japan, with 4,000+ members mainly based in Tokyo (engineers, researchers, investors, product managers, and corporate innovation managers).

​​​Value Create is a management advisory and corporate value design firm offering services such as business consulting, education, corporate communications, and investment support to help companies and individuals unlock their full potential and drive sustainable growth.

​Aurora Solutions K.K. is a specialist consulting firm in CCP clearing, Collateral and Risk Management, Digital Regulatory Reporting (DRR), DLT, and Generative AI for financial institutions. They partner with banks, clearing houses, and market infrastructures to deliver end-to-end solutions, from idea to production, helping their clients navigate complex regulations, modernise legacy platforms, and harness emerging technologies to accelerate the delivery of innovative services.

​​​Privacy Policy

​​​We will process your email address for the purposes of event-related communications and ongoing newsletter communications. You may unsubscribe from the newsletter at any time. Further details on how we process personal data are available in our Privacy Policy.