QCon AI New York 2026 Program Committee Picks: Agent Identity, Context Costs, and Open-Source Infere

Who authorizes an agent’s actions? How should teams manage growing context costs? What changes when coding agents become part of everyday engineering work? The QCon AI New York 2026 program committee has selected six sessions they are looking forward to at the December 15–16 conference. Their picks examine these questions alongside open-source inference, agent standards, and audio AI.

Conference chair Hien Luu, an AI/ML leader, advisor, speaker, and author, joins committee members Wes Reisz, Technical Principal at Thoughtworks and creator of The InfoQ Podcast, and Faye Zhang, Staff Engineer at Google building autonomous agents for Gemini Enterprise, in sharing their selections.

Wes on Agent Identity and Context Costs

Wes Reisz selected When Software Becomes a User: Identity and Authorization for Agents in Production, a keynote by Nancy Wang, CTO at 1Password.

As agents take on more autonomous work, traditional identity and authorization models start to break down. Nancy will explore delegated authority, credential management, and securing agent interactions, foundational capabilities I believe we must get right to truly adopt agentic delivery.

Nancy’s keynote examines how authority is delegated across subagent chains and how agents can access the credentials they need without exposing secrets in their context.

Wes’s second choice is Make Each Token Count: Compression, Caching, and Retrieval for Production LLM Context, by Tejas Chopra, CEO and co-founder of Headroom Labs.

Token costs and latency are real constraints when building production agents. Tejas will explore compression, caching, and retrieval strategies, particularly token compression, which I believe will become a must-have capability in modern agentic systems.

Tejas will cover learned prompt compression, prefix-cache-aware context alignment, and compress-cache-retrieve pipelines, including failure modes that affect tool calling and agent trajectories.

Hien on Engineering Organizations and Open-Source Inference

Hien Luu selected The Future of the Engineering Org, by Peter Bell, founder and CTO of Gather.dev.

Drawing on conversations with engineering leaders at organizations including Anthropic, OpenAI, Shopify, and Coinbase, Peter will examine how engineering structures and practices are changing around coding agents. Topics include team size and rituals, hiring and assessing engineers, changes in engineering leadership, and the future of junior developers.

Hien’s second pick is Open Source Agentic Inference: Unlocking the Harness with vLLM, by Francisco Javier Arceo, Forward Deployed Engineer at Baseten and maintainer of the vLLM Agentic API project.

Francisco will follow a coding task from the harness through model generation, tool execution, and continuation using Agentic API, vLLM, and llm-d. The session examines where tools run, who owns context, and how serving infrastructure supports these workloads. It also explores how repeated inference and context growth affect cost, including why cost per completed task matters more than token price alone.

Faye on Agent Standards and Audio AI

Faye Zhang described her selections as "two talks at the intersection of frontier research and production engineering". Her first choice is Navigating Agentic Standards: What to Build On Now and What to Watch, by Kevin Kelley, Technical Lead for Standards at Anthropic.

She also highlighted how standards such as MCP, A2A, and OAuth can establish interoperable agent identity, delegated authorization, and trust boundaries without creating new security vulnerabilities. Kevin’s session organizes the standards landscape around questions such as which agent is acting, on whose behalf, and how it reaches tools and other agents. It distinguishes what has shipped from work that remains open.

Her second selection is Audio Processing with Large Language Models, a keynote by Tara Sainath, Distinguished Research Scientist at Google DeepMind and lead of the Gemini audio area.

Faye highlighted the questions raised by LLMs reshaping speech and audio modeling: unified multilingual representations, end-to-end versus cascaded architectures, and the latency–quality trade-offs required for real-time conversational intelligence. Tara’s keynote will discuss research and product developments at Google across speech recognition, generation, translation, and dialogue.

Early-Bird Pricing Changes November 10

QCon AI New York 2026 takes place December 15–16 at the Westin Jersey City Newport, one PATH stop from Lower Manhattan. The two-day conference focuses on production AI for senior software engineers, architects, and engineering leaders.

Early-bird pricing is $1,705 until November 10, after which the two-day pass rises to $1,790. Anyone planning to use their remaining 2026 learning and development budget can explore the QCon AI New York program and the "convince your manager" approval email on the conference website.