Speaker: Mohamed Jawad Askar Ali – Head of AI Platform, Home Team Science Agency (HTX)
Description
Explore the challenges and solutions involved in enabling self-service deployment of complex LLMs/VLMs within an organization’s own infrastructure. This session covers our platform’s approach to simplifying the user journey from experimentation to production, integrating MLOps, security, and resource management on-prem.
Key Takeaways:
• Bridges the gap between AI experimentation and production deployment, revealing the engineering challenges behind seemingly simple language model interactions
• Provides data leaders with strategic insights into memory and computational bottlenecks that impact performance at scale
• Demonstrates practical approaches to optimize infrastructure investments and maximize return from expensive computing resources
• Illustrates how thoughtful architecture decisions can dramatically improve throughput, reduce costs, and enable enterprise-grade AI applications
• Offers a balanced perspective connecting data science innovation with operational excellence for sustainable AI implementation
