The full schedule for QCon AI Boston 2026 is now live. Running June 1-2 at Boston University, the two-day program groups sessions around the engineering problems that come after the AI demo: getting agents into production, keeping inference affordable, making non-deterministic systems auditable, and rethinking how teams build software when AI is in the loop.
Program chair Eder Ignatowicz, Senior Principal Software Engineer and Architect @Red Hat AI, frames the program around the gap between a demo that impresses stakeholders and a system that holds up under production traffic, cost constraints, and audit.
Context engineering for agents
Agents often perform well during the experimentation phase, but things can quickly go wrong when they have to operate inside a real company's services, data, and processes. Two sessions cover what it takes to close that gap.
Context Engineering at LinkedIn: How We Built an Organizational Context Layer for AI Agents with MCP, by Ajay Prakash, Senior Staff Software Engineer @LinkedIn, covers how LinkedIn used the Model Context Protocol (MCP) to help coding agents work with internal services and frameworks rather than treat every company the same.
Beyond Prompting: Context Engineering for Production-Grade AI, by Ricardo Ferreira, Lead, Developer Relations @Redis, covers what goes into building production-grade AI applications beyond prompt iteration, including the data and retrieval context that shapes reliable LLM outputs.
Inference economics and infrastructure
For systems operating at the scale of modern enterprise companies, inference cost and latency shape the architecture early. Three sessions cover different angles of that problem.
Serving LLMs at Scale: The Hidden KV Cache Advantage, by Khawaja Shams, Co-Founder & CEO @Momento, covers KV cache as the hidden lever behind inference cost and performance, with direct impact on GPU utilization, throughput, and "Time to First Token".
Beyond the Prototype: Scaling Frame Agnostic AI Agent Infrastructure with Ray, by Deepak Chandramouli and Bhumik Thakkar, both Senior Software Engineers @Apple, covers the transition from a local notebook to a production-grade "Agent Engine" that serves large-scale web services.
From Fab To Token: The State Of The Market, by Jordan Nanos, Member of Technical Staff @SemiAnalysis, is a data-driven look at the physical and economic bottlenecks in today's AI infrastructure market, including the divergent strategies between traditional hyperscalers and specialized "Neoclouds".
Reliability, evaluation, and safety
Several sessions treat safety, evaluation, and trust as engineering work.
SafeChat: Building AI-Powered Safety Systems at Scale in a Real-Time Marketplace, by Bruna Pereira, Software Engineer @DoorDash, examines trust and safety in real-time marketplace interactions.
Adaptive Recommenders in the Real World: Inference, Evals, and System Design, by Mallika Rao, Engineering Leader @Netflix, covers building an adaptive recommendation engine in production, one that continuously learns and evolves rather than running as a deployed model.
Building Reusable Evaluation Frameworks for Agentic AI Products, by Susan Chang, Principal Data Scientist @Elastic, covers methods for evaluating AI agents, including an example of frameworks built for a user-facing agent system that has been in production for almost two years.
Zero Trust Agent Systems that Pass Audits and Still Ship, by Advait Patel, Senior Site Reliability Engineer @Broadcom, examines the challenges of running agentic systems within strict enterprise boundaries where security, compliance, and incident response are non-negotiable.
AI inside the developer workflow
The program also examines how AI is changing the software development lifecycle (SDLC) and the accompanying engineering roles.
AI First, Quality Always: Agentic SDLC Adoption Case Study, by Catherine Weeks, Engineering Director @Red Hat, shares a case study on adopting an agentic SDLC without letting the pressure to show productivity gains crowd out quality and trust.
Tuesday's opening keynote, The Five Stages of AI Maturity in Engineering Organizations: Where and Why Teams Get Stuck, by Lizzie Matusov, Co-Founder & CEO @Quotient and co-author of Research-Driven Engineering Leadership, looks at where engineering organizations typically stall on the path to AI maturity, and what keeps them there.
The full schedule is at boston.qcon.ai. Early bird pricing and team discounts are available.