Keynote Presentation:
“Generative Computing: A principled approach for building robust secure and efficient GenAI applications“
The emergence of increasingly powerful and capable large language models has led to a rapid evolution in the scale and sophistication of AI-powered applications. Over the last 12 months, we have seen generative AI apps evolve rapidly from simple chatbots and Q&A systems to more sophisticated assistants, agents, and even multi-agent systems. For a single end-user query, many of these applications utilize multiple iterative LLM calls, often to multiple models, interspersed with business logic and control flow. Yet, the programming paradigm for interacting with LLMs has largely remained rudimentary and based on highly brittle and laborious prompt engineering. This has resulted in systems built using massive hand crafted prompts that are highly error-prone, hard to maintain, insecure, and inefficient to execute.
In this session, we will introduce and describe a new generative computing paradigm that establishes a robust and well-defined programming model and software infrastructure for building LLM applications. Using real examples from enterprise use cases, we will show how generative computing enables a systematic approach to building AI agents, promoting better maintainability, security, efficiency (through LLM-software co-design), and quality (through the use of inference scaling techniques). We will also describe how we are enabling generative computing through a new programming library on top of IBM’s Granite models.
About the Speaker:
Sriram Raghavan is Vice President at IBM Research for AI (artificial intelligence). In this role, he leads a worldwide team of research scientists and engineers who are advancing the field of AI and accelerating its applications to the digital transformation of enterprises. Sriram is responsible for establishing and executing a wide-ranging research agenda that spans foundational and applied AI. He also has overall responsibility for the R&D portfolio and transfer of technology from IBM Research to IBM’s $25B+ software business. Prior to his current role, Sriram was the Director of the IBM Research Lab in India and the CTO for IBM in India/South Asia. Sriram began his career in IBM at the Almaden Research Center in San Jose, California, USA where he led a variety of research efforts in natural language processing, data management, and distributed systems. Sriram is an alumnus of Stanford University, USA and the Indian Institute of Technology, Chennai, India. He is a recipient of multiple IBM Corporate and IBM Research Accomplishment Awards and is on the technical advisory board member of several research organizations.