Shalini Sivasamy
New York Life
Fairfax, Virginia, United States
Actions
Shalini Sivasamy is a Senior AI & Machine Learning Engineer specializing in building scalable, enterprise-grade artificial intelligence solutions across financial services and cloud ecosystems. Currently contributing to advanced AI initiatives at leading institutions, she focuses on leveraging Generative AI, Agentic AI, and Retrieval-Augmented Generation (RAG) to transform complex business workflows and decision-making processes.
With deep expertise in machine learning, natural language processing, and intelligent document processing, Shalini has developed production-ready AI systems that enhance operational efficiency, reduce processing time, and unlock actionable insights from unstructured data. Her work spans end-to-end AI system design from model development and evaluation to deployment using cloud-native architectures.
A strong advocate for responsible and trustworthy AI, Shalini actively explores challenges related to model reliability, governance, and real-world deployment of large language models. She has authored peer-reviewed publications and contributed to academic and industry research, while also serving as a reviewer for international conferences and journals in AI and data science.
Beyond her technical contributions, Shalini is a recognized thought leader and speaker in AI innovation, frequently sharing insights on enterprise RAG systems, agentic workflows, and AI-driven transformation in financial institutions. Her unique ability to bridge traditional software engineering with cutting-edge AI makes her a key voice in advancing intelligent, scalable, and responsible AI solutions.
Links
Area of Expertise
Topics
Agentic AI in the Enterprise: Building Workflows That Are Scalable, Trustworthy, and Audit-Ready
Enterprise AI has moved beyond chatbots and co-pilots. Agentic AI systems now autonomously orchestrate multi-step workflows — ingesting documents, extracting entities, validating outputs, and escalating to humans when confidence thresholds aren't met.
But autonomy without governance is a liability, especially in regulated industries.
This session examines real-world agentic AI architectures deployed in financial services — covering multi-agent pipeline design, human-in-the-loop escalation patterns, PII handling, hallucination detection, and audit-ready output generation. The focus is on what it takes to move agentic AI from impressive demo to production-grade enterprise system that compliance teams, auditors, and leadership can trust.
Practical takeaways for architects, engineers, and business leaders thinking about where agentic AI fits — and where it still needs a human in the loop.
Beyond RAG: A Compliance-Aware Framework for Regulated Enterprise AI
Enterprise AI deployments in banking and insurance often struggle with challenges that traditional RAG architectures do not adequately address, including PII leakage across retrieved documents, hallucinated responses in compliance-sensitive workflows, inconsistent schemas across legacy systems, and limited auditability for downstream decisions.
This session presents an intent-aware retrieval framework developed from real-world financial services deployments, where user queries are classified before retrieval and routed through specialized pipelines based on business intent, risk level, and compliance requirements. The architecture incorporates guardrails, confidence scoring with human escalation, and multi-agent orchestration patterns to improve reliability in regulated environments.
The framework introduces a six-stage pipeline: multi-format field extraction using LLMs for unstructured documents and deterministic parsers for structured formats; LLM-driven field confirmation through interactive user validation; hybrid semantic retrieval via OpenSearch VectorDB combining dense and sparse retrieval for accurate field-level similarity; context-enriched LLM mapping that compares source and target schemas using retrieved semantic chunks; and structured output generation for downstream migration and validation workflows. Human-in-the-loop checkpoints at critical stages ensure that only validated field mappings progress through the pipeline, directly addressing the reliability gap in automated schema translation systems.
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top