QCon AI New York 2026 Program Committee Picks: Agent Identity, Context Costs, and Open-Source Infere
Who authorizes an agent’s actions, and how should teams manage growing context costs? The QCon AI New York 2026 program committee shares six session picks that tackle these questions, alongside open-source inference, agent standards, and audio AI. By Artenisa Chatziou
The program committee for QCon AI New York 2026 has chosen six sessions for their attention at the upcoming conference on December 15-16. The committee includes chair Hien Luu, as well as selection members Wes Reisz and Faye Zhang.
Wes Reisz highlights a talk on agent identity and authorization for production - a topic of CTO Nancy Wang's keynote. Nancy will explore issues around delegation, credential management, and securing agent interactions as the need for agentic delivery grows.
Tejas Chopra's talk on token compression, caching, and retrieval for production LLM context is another key pick. As token costs and latency become real constraints, Tejas will cover compression techniques, learned prompt compression, prefix cache-aware context alignment, and compress-cache-retrieve pipelines.
Hien Luu selects Peter Bell's talk on the future of engineering orgs around coding agents. Drawing on discussions with leaders from Anthropic, OpenAI, Shopify, and Coinbase, Peter will examine how engineering structures and practices are evolving to accommodate agents. Topics include team size and rituals, hiring, leadership changes, and the future of junior developers.
Faye Zhang chooses two talks on standards and audio AI. For standards, she highlights Kevin Kelley's session on navigating agentic standards from Anthropic. The talk will explore questions around agent identity, delegated authorization, and trust boundaries, while distinguishing what is shipped from what remains open. For audio AI, Faye selects Tara Sainath's keynote on large language model applications to speech and audio.
Tara will discuss research and product developments at Google across speech recognition, generation, translation, and dialogue.
Written by urgent.news from InfoQ's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
