Annie Talvasto
CNCF Ambassador & Sr. Manager at Upbound
New York City, New York, United States
Actions
Annie Talvasto is an award-winning international technology speaker and leader who has spoken at 100+ events worldwide, including KubeCon + CloudNativeCon and Microsoft Build & Ignite. She has been recognized as a CNCF Ambassador and with Azure & AI Platform MVP awards. She co-founded Kubernetes & CNCF Finland in 2017 and has served as a KubeCon + CloudNativeCon track chair, Secure AI Summit Program Chair, and host of CNCF’s Cloud Native Live.
Links
Badges
Area of Expertise
Topics
The Past, Present, and Future of AI Infrastructure
Everyone says they know where AI is going. The truth is, no one really does.
So instead of pretending to predict the future, let’s look at the past. This keynote traces the relationship between AI and infrastructure through the people, ideas, and technical shifts that brought us here, from early machine intelligence to the platform shifts that shaped how we build and run systems today.
Along the way, we’ll ask what actually changes when AI meets production reality, and look at the tooling, patterns, and open source work platform teams can use right now. What patterns from cloud native history are repeating? What mistakes should we avoid? And what kind of infrastructure will this next era require?
The future is not predetermined. Especially in open source, it is something we get to build together.
Fable-Level Intelligence, Built in the Open
What if getting frontier-level AI capabilities did not mean sending every problem to the biggest proprietary model you can find?
Open models are getting remarkably capable. But the model is only part of the system. Tool use, context management, retrieval, routing, memory, verification, and agent harnesses can matter just as much as the name printed on the model card.
In this demo-heavy session, we will see how far we can get using open models and open source tools to build a system that tackles the kind of complex, multi-step work normally associated with frontier models. We will experiment with different models, give them tools, route work between them, and add mechanisms for checking and improving their own output.
The goal is not to claim that every open model magically matches Fable. It is to find out how much intelligence we can assemble from the pieces we can actually inspect, run, change, and own.
From Platform Engineer to Inference Engineer: You Already Know More Than You Think
If you know how to run applications on Kubernetes, you are already surprisingly close to understanding AI inference. Then someone says “tensor parallelism,” “KV cache,” or “prefill/decode disaggregation,” and suddenly it feels like you need another degree.
You don't.
This talk maps the world platform engineers already know to the inference stack they are being asked to operate. Scheduling becomes GPU scheduling. Resource requests become accelerator topology. Application replicas become hundreds of gigabytes of model weights. Autoscaling still exists, except cold starts can take minutes and state suddenly matters.
We will build an inference service from familiar Kubernetes primitives, then introduce the concepts that are actually new: models, tokens, accelerators, inference engines, batching, parallelism, routing, and the metrics that tell you whether any of it is working.
Come as a platform engineer. Leave speaking enough inference engineer to be dangerous.
AI Inference Best Practices: What Actually Works in Production
Running a model is easy. Running inference well is a lot harder.
What should you actually monitor? When should you batch requests? How do you choose an inference engine? Should you quantize? Scale up or scale out? How do you keep expensive GPUs busy without destroying latency? And what changes once you have more than one model, cluster, or type of accelerator?
This session walks through the practical best practices for running AI inference in production, from model and engine selection to GPU utilization, batching, caching, parallelism, routing, autoscaling, observability, and cost. We will look at the patterns that work, the common mistakes that look reasonable but hurt you later, and where the right answer depends on your workload.
With demos throughout, you will leave with a concrete checklist for designing, running, and improving a production inference stack.
Cloud Native AI Summit - EUROPE 2026 Sessionize Event Upcoming
KCD Istanbul 2026 Sessionize Event
Nerdearla Chile 2026 Sessionize Event
KubeCrash Fall 2025 Sessionize Event
KCD Helsinki 2025 Sessionize Event
CNCF-hosted Co-located Events Europe 2025 Sessionize Event
Nerdearla México 2024 Sessionize Event
swampUP 2024 Sessionize Event
dotnetdays 2024 Sessionize Event
DevOpsDays Zurich 2024 Sessionize Event
KubeCon + CloudNativeCon Europe 2024 Sessionize Event
NDC London 2024 Sessionize Event
KubeCon + CloudNativeCon North America 2023 Sessionize Event
Swiss Cloud Native Day 2023 Sessionize Event
.NET Day Switzerland 2023 Sessionize Event
Copenhagen Developers Festival 2023 Sessionize Event
Kubernetes Community Days Munich 2023 Sessionize Event
NDC London 2023 Sessionize Event
DDD North 2022 Sessionize Event
TechBash 2022 Sessionize Event
NDC Sydney 2022 Sessionize Event
NDC Oslo 2022 Sessionize Event
Swetugg Gothenburg 2022 Sessionize Event
.NET Day Switzerland 2022 Sessionize Event
KCDC 2022 Sessionize Event
Cloud Lunch and Learn Sessionize Event
Azure Bootcamp Switzerland 2022 Sessionize Event
NDC Porto 2022 Sessionize Event
CloudWorld 2022 Sessionize Event
Azure Developer Community Day 2021 (hybrid Edition) Sessionize Event
DDD 2021 Sessionize Event
vOpen 2021 Sessionize Event
Update Conference Prague 2021 Sessionize Event
Build Stuff 2021 Lithuania Sessionize Event
Automation + DevOps Summit Sessionize Event
Azure Community Conference 2021 Sessionize Event
2021 All Day DevOps Sessionize Event
API World 2021 Sessionize Event
TechBash 2021 Sessionize Event
PowerShell, DevOps and Cloud Conference Sessionize Event
KCDC 2021 Sessionize Event
Azure Summit Sessionize Event
Techorama 2021 Spring Edition Sessionize Event
Future Tech 2021 Sessionize Event
TheCloudFirst Sessionize Event
Azure Lowlands Sessionize Event
Annie Talvasto
CNCF Ambassador & Sr. Manager at Upbound
New York City, New York, United States
Links
Actions
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top