Optimizing LLM Applications

​Description:
The rapid evolution of Large Language Models (LLMs) presents significant opportunities alongside challenges such as fragmentation, high costs, and operational complexity. This presentation explores state-of-the-art strategies and technologies designed to optimize LLM deployments and streamline LLMOps. We will review programmatic development frameworks for prompt optimization, unified access through LLM gateways and intelligent model selection via LLM routing and orchestration. This talk aims to provide developers and enterprises with insights into building robust, cost-effective and high-performing LLM applications.

​About the speaker:
Andro Sabashvili, PhD is working on building LLM-powered agentic applications and scaling sales forecasting at Majid Al Futtaim. Previously, he led a data science team developing an automated machine learning platform utilizing his patent-pending optimization algorithm.

​Location: Rooms Hotel Tbilisi, Central Room -1 floor (The main entrance, Chovelidze street side)