AI-Augmented CRM Automation: Intelligent Enterprise Systems with Salesforce, LLMs, and Cloud-Native Architectures
United States
INTERNATIONAL DISTINGUISHED RESEARCHER AWARD
I’m a technology professional with close to ten years of experience working as a Senior Salesforce Architect and Full-Stack Engineer, focused on building and scaling large enterprise applications. Over the years, I’ve worked extensively across enterprise architecture, modern web and mobile frameworks, and distributed systems, with a consistent emphasis on performance, scalability, and audit readiness. More recently, my focus has shifted toward Natural Language Processing (NLP), Large Language Models (LLMs), and Generative AI engineering. I spend a lot of time exploring how AI architectures can move beyond theory and experimentation and be applied in practical ways within mobile, cloud, and enterprise systems. What interests me most is finding the right balance—combining traditional, proven software engineering practices with modern AI capabilities in a way that remains reliable, traceable, and suitable for enterprise environments. A significant part of my career has been spent designing and scaling mission-critical systems. Most recently, I led the architecture and development of a Risk Control and Self-Assessment (RCSA) platform at USAA. In that role, I designed and implemented a high-performance, compliance-ready solution using Salesforce, JavaScript, and modern UI frameworks, ensuring it met strict regulatory, audit, and scalability requirements. Working so closely with regulated systems naturally pushed me to explore how Generative AI could be applied responsibly in areas like risk management, controls, and compliance. Through this work, I developed a strong understanding of where AI can realistically add value to RCSA processes. My ongoing research looks at how LLMs can learn from historical RCSA data to suggest more consistent ratings and descriptions, assist with risk identification from process narratives, support risk and control taxonomy mapping across organizations, and provide guidance on control design. A key theme across all of this is the use of human-in-the-loop approaches, where AI supports decision-making while preserving governance, explainability, and auditability. Alongside my enterprise work, I remain deeply hands-on as an engineer. I regularly work across Java, JavaScript/TypeScript, Angular, React, React Native, Node.js, and Firebase, which allows me to design and build solutions end to end. I actively apply these skills in AI-driven mobile and productivity applications, where I experiment with natural-language reminders, time normalization, fuzzy language handling, entity extraction, semantic parsing, and multi-step reasoning for symbolic and mathematical problems. My AI learning and experimentation span the full lifecycle, starting from transformer fundamentals such as attention mechanisms and embeddings, all the way through tokenizer behavior, inference optimization, and fine-tuning strategies. I work hands-on with techniques like LoRA and QLoRA, instruction tuning, domain-specific dataset curation, and low-resource training workflows. I place a strong emphasis on understanding how transformers work internally and validating those concepts using real data, real user behavior, and real-world constraints. I’m also actively involved in building AI-ready mobile architectures using React Native, Expo, Node.js, and Firebase, where I compare open-source and custom model approaches and design AI-powered user experiences centered around conversational and contextual interaction. My overall approach to AI is incremental and data-driven—introducing structured extraction, tagging, contextual understanding, and learning capabilities step by step to maintain reliability and user trust. Looking ahead, my long-term goal is to build domain-specific language models that combine deep process understanding with contextual inference to power intelligent enterprise and mobile applications. I’m especially interested in AI systems that support learning, productivity, and everyday decision-making, with a strong focus on clarity, consistency, and responsible adoption. I aim to continue studying AI both theoretically and practically, diving deep into transformer architectures while constantly testing those ideas through hands-on engineering and applied experimentation.