Jason's AWS re:Invent 2026 Schedule
November 29 – December 4, 2026 • Las Vegas, NV • Day-by-Day Itinerary
14 Reserved
3 Walk-ups
1 Conflict
No schedule items match your search and filter criteria.
Sunday
November 29, 2026
20:20 – 21:00
Arrival Flight — Landing in Las Vegas (LAS)
Monday
November 30, 2026
Caesars Forum
1 session
07:00 – 10:30
07:00 – 08:30
90 min
Breakfast
Caesars Forum • Forum Food Court & Restaurant Row
08:30 – 10:30
2 hrs
AMZ202-R
Reserved
Workshop
Level 200
Turn a Bee wearable into your re:Invent notetaking companion
re:Invent keynotes don't slow down for note-taking. Last year's CEO keynote announced roughly 28 launches in two hours. Careful notes still leave you with names, not links. So build an agent that does the finding. Participants in this workshop will get a Bee Pioneer wearable: it transcribes while you listen, and a double-press bookmarks anything worth a deep dive. Your agent (Strands Agents on Amazon Bedrock) reads the transcript and researches each announcement through the AWS MCP Server, producing one markdown file linking every launch to its docs. Then make it yours: benchmark cheaper models, post to Slack, or run it in the cloud.
Caesars ForumLevel 1Forum 101Kourtney Meiss, Emerson Sklar
10:30 – 11:00
Walk: Caesars Forum → Caesars Palace (5 min + buffer)
Caesars Palace
2 sessions
11:00 – 14:30
11:00 – 12:00
1 hr
Lunch
Caesars Palace • Palace Court or Forum Food Court
12:00 – 13:30
90 min
TNC329-R
Reserved
Hands-on Lab
Level 300
Integrate Agents with MCP Servers and Tools using AgentCore Gateway
Your agent may handle basic conversations well, but to take real-world actions it needs access to external tools and APIs. In this hands-on lab, use Amazon Bedrock AgentCore Gateway to expose Lambda functions as MCP-compatible tools with JWT authorization. Register tool targets, connect to a pre-deployed nutrition coach agent, and validate how it dynamically discovers and invokes new capabilities at runtime without redeployment.
Caesars PalaceEmperors LevelPalace Ballroom IIMarco Tamassia
13:30 – 14:30
1 hr
HMC319
Reserved
Chalk Talk
Level 300
Migrate and modernize workloads with AWS Outposts
Organizations running on-premises workloads on legacy virtualization platforms are looking to migrate applications to AWS and modernize them to be more resilient and operationally consistent with their cloud workloads. AWS Outposts supports customers on this journey and is now available in second-generation single- and multi-rack offerings that deliver new instance types and Amazon EBS storage, up to 40% better price-performance, and enhanced networking features for monitoring and security. This session covers migration and modernization strategies for organizations looking to move to a cloud operating model on-premises, as well as strategies for first-generation Outposts customers to upgrade to the latest generation.
Caesars PalacePromenade SouthOctavius 9Vaibhav Phadnis, Brianna Rosentrater
14:30 – 15:30
Transit: Caesars Palace → Wynn/Encore (15 min walk or shuttle)
Wynn/Encore
1 session
15:30 – 17:30
15:30 – 17:30
2 hrs
ARC331
Reserved
Workshop
Level 300
SaaS Agents Anywhere: building distributed agentic deployments
SaaS vendors deploy agents across a spectrum: fully provider-hosted, fully tenant-controlled, and hybrid models in between. Once customers have data residency or cost-visibility requirements, deployment topology determines the operational pattern. In this workshop, build a distributed agentic deployment on AWS spanning your accounts and a tenant's, then exercise the three concerns that surface everywhere on the spectrum: lifecycle management of agents you can't directly see, observability across account boundaries, and reliability when the agent runs somewhere you don't operate. Leave with a working reference deployment you can adapt to your own tenancy model.
Wynn/EncoreLevel 1Encore Ballroom 1Hasan Tariq, Markus Doepke
Tuesday
December 1, 2026
Wynn/Encore
1 item
07:00 – 08:30
07:00 – 08:30
90 min
Breakfast
Wynn/Encore • Buffet at Wynn or Terrace Pointe Café
Venetian
1 session
08:30 – 10:30
08:30 – 10:30
2 hrs
KEY002
Walk-up
Keynote
CEO Keynote with Matt Garman
Join AWS CEO Matt Garman for the re:Invent 2026 CEO Keynote as he shares the latest innovations powering the agentic era. From AI infrastructure and inference to data, agent platforms, and the software agents AWS is building, see how AWS gives you the depth, breadth, and integration to build what's next with new announcements and stories from customers already doing it.
VenetianLevel 2Ballroom EMatt Garman, Barbara Karuth-Zelle, Tony Xu
Wynn/Encore
1 session
11:30 – 14:30
11:30 – 12:30
1 hr
Lunch
Wynn/Encore • Wynn Food Court or Convention Area Dining
12:30 – 14:30
2 hrs
GHJ320
Reserved
Gamified Learning
Level 300
re:Architecture Rodeo: Building an AI-Ready Data Foundation
Your data estate wasn't designed for AI agents. Nobody's was. But somewhere between your RDBMS, NoSQL clusters, and that untouchable data lake, an agentic-ready architecture waits to be built. In this Rodeo session, your team draws a challenge: a legacy warehouse needing a semantic layer, a sprawling lake needing an ontology, or a metadata mess that lets agents in without letting governance out. Your job: build a data layer that makes agents productive, keeps compliance calm, and gives analysts self-service. You'll whiteboard ontologies, debate metadata, and wire up multi-agent access against the clock. Best architecture rides away with the win.
Wynn/EncoreConvention PromenadeLatour 2Jigna Gandhi, Anurag Singh
14:30 – 16:00
Break & Open Exploration Window
Wynn/Encore
2 sessions
16:00 – 17:00
16:00 – 17:00
1 hr
COP304
Walk-up
Breakout Session
Level 300
Observability and operations in the AI era
Generative AI broke the assumptions traditional monitoring was built on, and teams struggle to keep non-deterministic systems predictable, explainable, and affordable. This session opens on that challenge, then shows how the newest observability capabilities give you visibility into model quality, cost, and agent behavior across your AI stack, using traces, spans, and confidence scoring, illustrated with real customer patterns.
Wynn/EncoreLevel 1Encore Ballroom 5Andres Silva (AWS), Pratul Chakre
16:00 – 17:00
1 hr
ARC402
Walk-up
Builders' Session
Level 400
Close the loop: From guardrail blocks into layered defense
Guardrails block malicious prompts. Every blocked attempt is a signal: proof of what your defenses caught and a clue toward what will be attempted next. Capture that signal instead of discarding it. In this session, you design and deploy a layered enforcement architecture that closes the loop. A stateful detection tier identifies repeated probing across sessions and propagates block decisions to the network edge. You evaluate signals from guardrail violations and apply policies to classify and block the threat. You leave with a deployed pattern applicable to any AI agent workload. Bring your laptop to participate.
Wynn/EncoreUpper Convention PromenadeCristal 2Jonathan Woods, Patrick Duffy, Wei Chen, Jason Schamp, Sanghwa Na
Wednesday
December 2, 2026
MGM Grand
3 sessions
07:00 – 14:30
07:00 – 08:00
1 hr
Breakfast
MGM Grand • MGM Grand Buffet or Conference Center Grab & Go
08:00 – 10:00
2 hrs
DAT429
Reserved
Workshop
Level 400
Multi-tenant agentic AI with Aurora and DynamoDB
AI agents are autonomous, stateful, and capable of invoking external tools. In a multi-tenant SaaS system, every one of those capabilities is a potential data isolation failure. In this workshop, build a complete multi-tenant agentic AI solution enforcing tenant isolation across Amazon Aurora PostgreSQL with row-level security, Amazon DynamoDB, and Amazon S3 Vectors. Implement tenant-scoped vector stores for RAG, agent memory backed by multiple database engines, MCP-based tool authorization gated by subscription tier, and usage metering. Leave with a working reference architecture that prevents cross-tenant data leakage. Bring a laptop to participate.
MGM GrandLevel 3Premier 314Arjan Schaaf, David Gardner
10:00 – 11:00
1 hr
NET329-R
Reserved
Chalk Talk
Level 300
Scaling and securing MCP Servers with AWS Load Balancing and WAF
MCP servers are becoming essential infrastructure in enterprise AI environments, and running them in production demands both scalability and security. Join this chalk talk to explore patterns for scaling and securing your MCP servers using AWS native services. We'll cover how to configure Application Load Balancer for MCP traffic characteristics, implement health checks and routing policies, and scale horizontally across availability zones. We'll also cover how to enforce security controls with AWS WAF, including request filtering, rate limiting, and tool-level access policies. Whether you're running a handful of MCP servers or deploying at enterprise scale, you'll leave with actionable architectures for production-ready deployments.
MGM GrandLevel 1Boulevard 163Max Hausner, Eitav Arditti
11:00 – 12:30
90 min
Lunch
MGM Grand • Conference Center Food Court or Studio Café
12:30 – 14:30
2 hrs
ANT416-R1
Reserved
Workshop
Level 400
Build agentic AI applications with the open data lakehouse architecture on AWS
Transform your organization's data landscape and eliminate silos in this hands-on workshop with the open data lakehouse architecture on AWS. Unify data management and design data interoperability across data lakes, data warehouses, and federated data sources. Through practical exercises, discover how to create a single source of truth while maintaining security and governance across your entire data estate. You'll implement real-time analytics, optimize query performance, and design scalable architectures. Perfect for data engineers, ML practitioners, and solution architects looking to modernize their data infrastructure for building agentic AI applications.
MGM GrandLevel 3Premier 313David Greenshtein, Nathan Yee
14:30
Walk: MGM Grand → Caesars Forum (10 min via elevated walkway — tight transition)
Caesars Forum
1 session
14:30 – 15:30
14:30 – 15:30
1 hr
SVS320
Reserved
Builders' Session
Level 300
Same code, any compute: Portable architecture on AWS
Your business logic shouldn't be coupled to a compute service. In this session, apply Hexagonal Architecture and 12-factor patterns to design workloads where compute is a deployment decision, not an architectural one. Through live code, see the same logic packaged as a Lambda function, a Fargate task, and an EKS pod using AWS Lambda Web Adapter and runtime-neutral handler patterns. You'll learn where portability pays off, where it adds unnecessary abstraction, and how to evaluate the trade-offs between pay-per-request, always-on containers, and orchestrated clusters for your workloads.
Caesars ForumLevel 1Forum 120 | Content Hub | Builders' Session 2Nicolas Tarzia, Daniel Abib, Swarna Kunnath, Michael Costa, Rafael Barbosa
15:30 – 16:00
Walk: Caesars Forum → Caesars Palace (5 min via connector bridge)
Caesars Palace
1 session
16:00 – 17:30
16:00 – 17:30
90 min
TNC335-R1
Reserved
Hands-on Lab
Level 300
Securing AI Agents on AgentCore
A customer runs a support agent on Amazon Bedrock AgentCore that reaches destinations nobody reviewed and takes requests from any caller. In this SimuLearn session, you work the case as their security expert. Behind Amazon API Gateway, AWS WAF Bot Control labels each script and AI crawler, and you write the rule that refuses them with a 403 that names the reason. On the way out, AWS Network Firewall tags every destination by content category. You watch in alert mode, then reject AI and search categories while permitted sites still load. You leave knowing how to classify, then control, what enters and leaves an agent.
Caesars PalaceEmperors LevelAugustus Ballroom VPrabhu G, Joe Huang
Thursday
December 3, 2026
Caesars Forum
1 session
07:00 – 10:30
07:00 – 08:30
90 min
Breakfast
Caesars Forum • Forum Food Court & Restaurant Row
08:30 – 10:30
2 hrs
SVS304-R1
Reserved
Workshop
Level 300
Building event-driven architectures
Event-driven architectures help development teams operate more independently, release features faster, and build more scalable applications. In this workshop, learn the fundamentals of event-driven design using Amazon SNS, Amazon SQS, AWS Lambda, Amazon EventBridge, and more. Through hands-on exercises, discover how to choose the right AWS messaging and eventing service for your use case, design loosely coupled systems that scale independently, and optimize for both cost and performance. Whether you're starting your event-driven journey or refining existing patterns, you'll leave with practical skills you can apply immediately. You must bring your laptop to participate.
Caesars ForumLevel 1Forum 101Katherine Vishnevskaya, Geoffroy Weyer
10:30 – 11:30
Walk: Caesars Forum → MGM Grand (10 min walk)
MGM Grand
1 session
11:30 – 13:00
11:30 – 12:00
30 min
Lunch
MGM Grand • MGM Grand Food Court
12:00 – 13:00
1 hr
ANT410
Reserved
Chalk Talk
Level 400
Give your AI agent an expert-level playbook with the Agent Toolkit for AWS
What if your AI agent could explore schemas, generate correct SQL, load data, and run migrations — all through the Agent Toolkit for AWS with expert-level context? In this chalk talk, we explore how the AWS MCP Server paired with packaged AWS Analytics skills give agents like Claude Code, Kiro, and Cursor deep AWS service expertise — from SQL syntax and metadata discovery to migration planning and performance tuning. Bring your toughest questions — we'll cover setup, agent workflows, and best practices for your environments.
MGM GrandLevel 3Room 353Avichay Marciano, Brandon Schur
13:00
Zero Buffer Alert: MGM Grand → Caesars Forum (10 min travel needed)
Caesars Forum
2 sessions
13:00 – 17:00
13:00 – 14:00
1 hr
SVS341-R
Reserved
Chalk Talk
Level 300
Lambda performance tuning and cost optimization
Most Lambda functions run at default memory settings, which means you're paying more for slower execution. Right-sizing memory often reduces cost and latency at the same time. In this chalk talk, we go beyond memory tuning: cold start elimination with SnapStart and provisioned concurrency, architecture-level savings like batching and right-sizing event source mappings, and when Lambda Managed Instances deliver better economics than per-invocation pricing. We whiteboard the decision points using real cost data and discuss which CloudWatch and Cost Explorer metrics actually drive optimization decisions.
Caesars ForumLevel 1Forum 103Matt Diamond, Joe Losinski
Starts immediately after ANT410 at MGM Grand with zero travel buffer.
14:00 – 16:00
2 hrs
Extended Break & Networking
Caesars Forum • Various networking areas & food stations
16:00 – 17:00
1 hr
SVS332-R
Reserved
Chalk Talk
Level 300
Lambda Managed Instances: Always-on Serverless Compute
Some Lambda workloads run constantly. For steady-state, high-throughput functions, per-invocation pricing becomes more expensive than running the same code on EC2, but you don't want to manage instances. Lambda Managed Instances give you EC2 economics with multi-concurrency while Lambda handles instance lifecycle, patching, and scaling. In this chalk talk, we whiteboard the decision: when Managed Instances beat standard Lambda pricing, what workloads migrate cleanly versus what needs adaptation, and how to model your break-even point. Bring your Lambda bill. We'll help you find where Managed Instances save money and where they don't.
Caesars ForumLevel 1Academy 417Nihar Sheth, Pushparaju Thangavel
Friday
December 4, 2026
06:05 – 13:20
Departure Flight: LAS → CLT (Terminal 1)
13:20 – 15:26
Layover in Charlotte (CLT)
15:26 – 17:39
Connecting Flight: CLT → BOS (Terminal B)
Venues
Caesars Forum
Caesars Palace
MGM Grand
Wynn/Encore
Venetian
✓ Reserved
Walk-up Option
Meal / Break
Walking / Transport
HIGHLIGHT TEXT
New Session Note
Unlock Note Editing
Viewing notes uses your schedule password. Enter your Notes Write Passphrase to add, edit, or delete notes and highlights in this browser session.