Agentic AI Takes Center Stage at Numerix and AWS Hosted Event

Numerix and Amazon Web Services (AWS) recently hosted an exclusive event in New York City, bringing together industry leaders to explore the future of capital markets. The evening was filled with insightful discussions and a compelling live demonstration, all centered on how AI is actively reshaping financial services. 

In today’s blog, we take a look at the pivotal moments and insights that defined the evening’s discussions.

Transforming Financial Services with Generative AI

The centerpiece of the event was a fireside chat featuring Numerix's Chief Information Officer, Nestor Nelson, and AWS's Financial Services Specialist, Alvin Huang. Moderated by Numerix CMO James Jockle, the discussion explored the rapid evolution of AI from simple chatbots to sophisticated, autonomous systems that can handle complex financial workflows.

Alvin Huang kicked off the discussion by noting the significant shift in AI conversations over just the past two years. The focus has moved from internal knowledge retrieval using Generative AI to deploying Agentic AI for production use cases. Agentic AI refers to systems that can operate with a high degree of autonomy, making decisions and taking actions with minimal human oversight. This evolution, he explained, is unlocking new efficiencies in areas like compliance, research, and report generation.

Nestor Nelson shared how Numerix is navigating this evolution. He described the journey from initial chatbot usage to exploring enterprise-grade coding assistance, automated testing, and product feature expansion. A key challenge, Nelson pointed out, is creating a consistent data domain model. Terms like "yield curve" can have different meanings depending on the context. He emphasized the importance of establishing a clear data strategy and governance framework to ensure AI models are trustworthy and compliant. Nelson also touched on Numerix's work in building a Hedge Backtesting Framework to accelerate collaboration with clients on model calibration.

The speakers agreed that governance is non-negotiable. Huang used an actual client example to illustrate how firms are tackling challenges like model hallucinations and building trust in AI-generated outputs. Nelson detailed Numerix's approach, which aligns with the NIST AI Risk Management Framework, establishing clear policies on AI usage and data access to ensure security and compliance.

A Live Look at Agentic AI in Portfolio Management

Following the fireside chat, Kris Skrinak, a Financial Service AI thought leader at AWS, delivered a featured session on "Dynamic Asset Allocation with Agentic AI." This live demonstration was a powerful illustration of the concepts discussed earlier, showing how AI agents can revolutionize portfolio optimization.

The session addressed a common limitation in portfolio management: the time-consuming and manual process of hypothesis testing. Traditionally, a portfolio manager might evaluate only five to ten strategies per quarter, a process that can take weeks for each strategy.

Skrinak showcased an "Intelligent Financial Loop" where AI agents, integrated with Numerix's CrossAsset analytics via its SDK, automated the entire workflow. The system autonomously generated and tested over 1,000 strategy configurations in minutes. The demonstration highlighted how agents could explore a vast range of hyperparameters—including target volatility, lookback windows, and rebalancing frequencies—across multiple market stress scenarios.

The process involved a series of specialized agents:

• A Strategy Agent generated potential configurations based on objectives defined by the portfolio manager.
• The Numerix CrossAsset SDK ran computationally intensive Monte Carlo simulations via an API.
• An Analysis Agent evaluated the risk-return characteristics of each simulation.
• An Optimization Agent identified the optimal parameters.
• Finally, a Recommendation Agent presented the findings with clear explanations.

The results were striking. The AI-driven approach delivered measurable improvements in Sharpe ratios and drawdown reduction, uncovering optimal allocations that traditional methods would likely miss. This demonstration proved that Agentic AI is not just a theoretical concept but a practical tool that can deliver a significant competitive edge and exceptional ROI.

The Future is Collaborative and Agentic

The evening concluded with a shared vision for the future. Both Nelson and Huang highlighted Agentic AI as a key innovation that will reshape financial services. Nelson predicted that agentic workflows will become the successor to traditional APIs, enabling SaaS providers like Numerix to meet clients' evolving needs for tighter integration and automation.

The collaboration between Numerix and AWS is already bearing fruit. Nelson highlighted ongoing work to build observability solutions using CloudWatch Insights and developing a natural language front-end for the NxCore Analytics Services SaaS offering.

The insights shared throughout the evening made one thing clear: the era of Agentic AI in finance is here. By combining advanced analytics from Numerix with the scalable power of AWS, financial institutions can move beyond manual processes and unlock new levels of efficiency, optimization, and innovation.