QCon AI Boston. July 8-9, 2027.

Once agents, retrieval, and inference run in production, a specific set of engineering decisions follows: who owns state when an agent runs for hours, how much access it gets to internal systems, what has to be visible before a change ships, and which requests need the most capable model. The QCon AI Boston 2027 program is being developed around those decisions. Those production examples give attendees reference points for comparing approaches in their own systems, including which constraints shaped a choice and what changed after launch. QCon AI Boston 2027 is a two-day, in-person conference on July 8-9, 2027, at Boston University, George Sherman Union. It is for senior engineers, architects, Staff+, ML, platform and security engineers, and engineering leaders responsible for AI inside production software systems. QCon hand-picks speakers for first-hand production experience with these questions. Sessions cover what a team considered, what it chose, what changed once the system was running, and what the practitioner would approach differently now. Sponsored sessions are identified separately from the editorial program, so there are no hidden product pitches. Between sessions, attendees can ask speakers why they rejected an approach, hear how another team handled a similar constraint, and continue the discussion with engineers working on related decisions at other companies. QCon AI applies the practitioner-led approach QCon and InfoQ have taken to software technology for more than 20 years to production AI. What You’ll Experience at QCon AI Boston 2027 Sessions that cover the alternatives a team considered, the trade-offs it accepted, what held up in production, and what the practitioner would change now. A program built around six production AI areas, from agent runtime design to inference cost and latency. Time to put follow-up questions to speakers and continue the technical discussion with senior practitioners from other companies. Sponsored sessions identified separately from the editorial program. On-demand access to most session recordings for 12 months, starting two weeks after the conference. Continental breakfast, lunch, coffee breaks, social events and receptions. Program Areas Agent Runtime Design Context Engineering Evals and Observability MCP and Tool Access Zero-Trust Agent Security Inference Cost and Latency Bring Your Team Architecture, platform, security, ML and engineering leadership can follow different parts of the program, compare what each person heard, and bring the same production examples back to the decisions they make together. Group discounts are available, with eligibility and savings depending on the booking date and team size. For current group and alumni terms, see the registration page or email [email protected] with your group size. Ticket prices increase at published deadlines. See the live pricing schedule for current prices and dates. For all other inquiries, please email [email protected].

Event Links

Tickets: https://go.evvnt.com/3867045-0

Website: https://go.evvnt.com/3867045-2

LinkedIn: https://go.evvnt.com/3867045-3

Twitter: https://go.evvnt.com/3867045-4

Instagram: https://go.evvnt.com/3867045-5

YouTube: https://go.evvnt.com/3867045-6

Read More

View Less