Title: PANEL DISCUSSION – FROM PROMPT TO PRODUCTION: OPERATIONALIZING AGENTIC LLM SYSTEMS

Speakers: Yubin Kim, Gautam Solaimalai, Shaleen Kumar Gupta, Vishal Jain, Abhay Khosla

Date: September 17th 2025, 6:00pm PT (Virtual)

Venue: Online/Zoom

Registration link : https://sjsu.zoom.us/meeting/register/i6n2sgjLQFelwXCNz4-YGQ

Organized by: IEEE Computational Intelligence Society Chapter of Silicon Valley (SCV), IEEE Computer Society, Santa Clara Valley Chapter

Co-sponsored by: Prof. Vishnu S. Pendyala, SJSU

Abstract:

As large language models (LLMs) evolve from static, prompt-based tools into autonomous, agentic systems capable of reasoning, planning, and acting with minimal human oversight, organizations face an exciting yet complex frontier. These advanced systems hold the potential to revolutionize enterprise workflows, developer tools, and customer-facing applications—but realizing that potential requires navigating a host of technical and ethical challenges.

This panel brings together leading voices from AI research, infrastructure engineering, and real-world application domains to discuss how agentic LLM systems are moving from lab experiments to production-grade deployments. Panelists will explore critical topics such as orchestration, safety, observability, and evaluation, while offering hard-earned lessons from deploying these systems at scale.

Whether you’re building tools for developers, integrating LLM agents into enterprise pipelines, or shaping the next wave of intelligent products, this discussion will equip you with the strategic and technical know-how to bring agentic AI into impactful, everyday use. Don’t miss this opportunity to learn what it truly takes to operationalize the future of AI.

Speaker bio:

Yubin Kim

Biography:

Dr. Yubin Kim is the Co-Founder and Chief Science Officer at Vody, a generative AI startup focused on transforming e-commerce discovery using multimodal large language models. At Vody, she leads scientific strategy and product development, building systems that enhance product search, recommendations, and catalog understanding by combining structured data with visual and textual signals.

Yubin holds a Ph.D. in Language Technologies from Carnegie Mellon University, where her research focused on information retrieval, distributed selective search, and retrieval system evaluation.  Before founding Vody, Dr. Kim held key applied machine learning roles in industry, including leading search and recommendation teams at Etsy, where she worked on improving buyer experience through personalized discovery systems. She has a track record of translating rigorous research into impactful, user-facing products.

In addition to her industry work, Yubin is active in the research community. She has served as a program committee member, organizer, and reviewer for top conferences such as SIGIR, CIKM, and ICTIR, and is a frequent speaker at workshops and forums bridging academia and industry.

Gautam Solaimalai

Biography:

Gautam Solaimalai is a Vice President and Senior Software Engineering Manager at U.S. Bancorp, specializing in cloud-native enterprise architecture, AI-driven automation, and financial technology innovation. With over 15 years of experience across industry leaders like Honeywell, OneTrust, and U.S. Bank, Gautam has led strategic product development and digital transformation initiatives that power secure, scalable financial platforms. He is a published author in IEEE journals, a peer reviewer for global tech conferences, and a recipient of multiple innovation and excellence awards. Gautam is also an active contributor to the technology community through mentoring, technical leadership, and applied research in AI, IoT, and DevOps.

Shaleen Kumar Gupta

Biography:

Shaleen works on Google Search’s AI Mode, focusing on post-training its multi-step, thinking and advanced reasoning LLM agent to enhance response quality and factuality. Previously, he made contributions to AI Overviews for search verticals. His past work involved training Multitasking Unified Model (MUM) based ranking models for search verticals and developing multimodal ranking for short videos and shopping. Throughout his time at Google, his efforts have centered on improving retrieval, ranking, and response generation for various search applications.

Vishal Jain

Biography:

Vishal Jain is a Staff software engineer at Meta and works on building scalable digital solutions. With extensive experience in quantitative finance and technology leadership roles at Meta, Two Sigma, and Bloomberg, he brings expertise in developing innovative solutions across industries. Vishal’s strong foundation in engineering and data-driven problem-solving, supported by his education at IIT Kanpur and Columbia University, positions him as a forward-thinking leader in the tech space.

Abhay Khosla

Biography:

Abhay Khosla is a Software Engineer at Google, specializing in large-scale distributed systems and applied machine learning. He has previously worked as a Machine Learning Engineer at Ocrolus, where he developed state-of-the-art computer vision models and benchmarking platforms, and as a Research Software Engineer at IBM Research, where he focused on natural language processing and knowledge graph population for AI-driven applications. Abhay holds an M.S. in Computer Science (Machine Learning Track) from Columbia University and a B.Tech. (Hons.) in Electronics and Electrical Communication Engineering with a Minor in Computer Science from IIT Kharagpur. His experience spans building scalable AI systems, deep learning research, and deploying production-grade ML services