Senior Engineer @Meta | IITD | RLHF, SOTA, Multi-Agent Systems, LLM Internals, Inference Scaling, Microservices Architecture | Distributed Engineering
2026 is the year AI agents move from demos to production, but most multi-agent systems still fail in the wild. This talk dives deep into the architectural patterns, orchestration frameworks, and engineering principles behind reliable agentic systems. We'll explore tool use, memory, planning loops, and inter-agent communication, alongside the often-overlooked challenges of observability, cost control, and failure recovery. Expect real-world case studies, anti-patterns to avoid, and a practical blueprint for building agents that ship.
AI voice is moving fast, but production is a very different test than a polished demo. Real-world voice applications have to handle latency, interruptions, call transfers, unreliable networks, legacy telecom systems, compliance requirements, and human escalation without breaking the experience. This roundtable will explore what it really takes to make AI voice work beyond the demo, from infrastructure and architecture to interoperability, reliability, and trust. Panelists will discuss where AI voice is working today, where it still falls short, and what builders need to solve next to make real-time AI communication ready for production.