Andrey Velichkevich
@andreyvelich.bsky.social
๐ค 1104
๐ฅ 17
๐ 15
Kubeflow Steering Committee | Work at Apple
JobSet scaling to 130,000 Pods across 130,000 Nodes โ sustaining 1,000 Pods/sec. Incredible to see
#Kubernetes
pushed to this level of performance ๐
cloud.google.com/blog/product...
loading . . .
How we built a 130,000-node GKE cluster | Google Cloud Blog
Learn about the architectural innovations we used to build a 130,000-node Kubernetes cluster, and the trends driving demand for these environments.
https://cloud.google.com/blog/products/containers-kubernetes/how-we-built-a-130000-node-gke-cluster/
4 months ago
0
2
0
Kubeflow Trainer v2.1.0 is released! โก๏ธ Stream in-memory tabular data to GPUs from distributed cache with zero-copy transfer ๐ฅ Fine-tune LLMs on
#Kubernetes
with MLX on CUDA โ now easier than ever! ๐ง Topology Aware Scheduling with
#Kueue
or
#Volcano
โ essential for GB200
bit.ly/4qO8iM1
loading . . .
#gb200 #gsoc #gsoc | Andrey Velichkevich
We have exciting news just in time of KubeCon + CloudNativeCon โ Kubeflow Trainer v2.1.0 is available! This release brings powerful new features, performance boosts, and enhanced flexibility for large...
https://bit.ly/4qO8iM1
4 months ago
0
1
0
Kubeflow Trainer 2.0 is here ๐ Built in collaboration with the
#Kubernetes
&
#Kubeflow
communities to make scalable AI model training easier than ever - with a Python SDK, resilient
@pytorch.org
support, LLM fine-tuning, gang scheduling, MPI runtimes & more.
blog.kubeflow.org/trainer/intro/
loading . . .
Democratizing AI Model Training on Kubernetes: Introducing Kubeflow Trainer V2
Running machine learning workloads on Kubernetes can be challenging. Distributed training and LLMs fine-tuning, in particular, involves managing multiple nodes, GPUs, large datasets, and fault toleran...
https://blog.kubeflow.org/trainer/intro/
8 months ago
0
2
0
Want to see how we've made it super easy to perform distributed training for ML frameworks like MLX and DeepSpeed on
#Kubernetes
? Check out our talk tomorrow at 2pm: From High Performance Computing To AI Workloads on Kubernetes: MPI Runtime in Kubeflow TrainJob
sched.co/1tx9k
loading . . .
KubeCon + CloudNativeCon Europe 2025: From High Performance Computing To AI Wo...
View more about this event at KubeCon + CloudNativeCon Europe 2025
https://sched.co/1tx9k
11 months ago
0
0
0
This
#KubeCon
+
#CloudNativeCon
2025 in London promises to be an inspiring and insightful event. Donโt miss these sessions to discover how weโre pushing the boundaries of innovation in Cloud Native AI/ML and
#GenAI
๐
sched.co/1tx9k
sched.co/1tcz0
sched.co/1u5fl
sched.co/1u5ii
12 months ago
0
1
0
reposted by
Andrey Velichkevich
Kubeflow Project
about 1 year ago
๐ New Kubeflow Python SDK Proposal! ๐ We're working on a Kubeflow Python SDK to improve the user experience for data scientists & ML engineers. More information can be found at:
groups.google.com/g/kubeflow-d...
We need your feedback!
#Kubeflow
#AI
#ML
#PythonSDK
loading . . .
[PROPOSAL] Kubeflow Python SDK
https://groups.google.com/g/kubeflow-discuss/c/cCWfvJ6EdCg/m/DAbKJQcREQAJ
0
2
2
Truly inspiring to see a student I mentored during GSoC 2024 presenting at
#KubeCon
+
#CloudNativeCon
๐ Kubeflow is a fantastic opportunity for anyone looking to shape the future of AI and cloud native LLMOps - donโt miss out!
youtu.be/4myE0DPp6Ko
loading . . .
Lightning Talk: Enhancing Hyperparameter Optimization with Advanced Parameter... - Shashank Mittal
YouTube video by CNCF [Cloud Native Computing Foundation]
https://youtu.be/4myE0DPp6Ko
about 1 year ago
0
1
0
Excited to introduce the new MPI Runtime in Kubeflow Trainer V2 at
#KubeCon
+
#CloudNativeCon
Europe. We will showcase how it empowers ML frameworks like MLX, DeepSpeed, and NVIDIA NeMo to streamline distributed AI model development on
#Kubernetes๐
sched.co/1tx9k
#AIML
#HPC
@cncf.bsky.social
loading . . .
KubeCon + CloudNativeCon Europe 2025: From High Performance Computing To AI Wo...
View more about this event at KubeCon + CloudNativeCon Europe 2025
https://sched.co/1tx9k
about 1 year ago
0
1
0
reposted by
Andrey Velichkevich
Kubeflow Project
about 1 year ago
๐ Exciting news from the Kubeflow community! Welcome Francisco Javier Arceo & Julius von Kohout to the Kubeflow Steering Committee! ๐ Huge thanks to Mathew Wicks, Josh Bottum, & James Wu for their leadership & dedication. More information:
groups.google.com/g/kubeflow-d...
#Kubeflow
#OpenSource
#AI
loading . . .
Welcome to Our New Kubeflow Steering Committee Members
https://groups.google.com/g/kubeflow-discuss/c/tTP7u0NydeM/m/xEtkaL6oBwAJ
0
2
2
It is incredible to see what our team has accomplished externally over the past year. I am super proud to be part of this journey. More things to come in 2025! ๐ฃ๏ธPublic talks:
lnkd.in/ebbEUU8X
๐ Leadership:
lnkd.in/ew6GXcex
๐ป OSS Contributions:
lnkd.in/ei2j2wrk
loading . . .
Sign Up | LinkedIn
500 million+ members | Manage your professional identity. Build and engage with your professional network. Access knowledge, insights and opportunities.
https://lnkd.in/ebbEUU8X
about 1 year ago
0
0
0
I am excited to join the Program Committee for Kubeflow Summit, co-located with
#KubeCon
+
#CloudNativeCon
in London 2025, alongside
@akgraner.bsky.social
. Have a story to share about
#Kubeflow
ecosystem and Cloud Native AI/ML? CFP is open by December 4th ๐
bit.ly/4ib9S6a
loading . . .
Kubeflow Summit | LF Events
Kubeflow is the MLOps platform of choice, used across the globe, by data scientists and machine learning engineers to develop and deploy models. It is a cloud-native application designed to run AI atโฆ
https://bit.ly/4ib9S6a
over 1 year ago
0
4
0
reposted by
Andrey Velichkevich
Chris Aniszczyk
over 1 year ago
๐
#KubeCon
#CloudNativeCon
Londonโs CFP hit record breaking with 2800+ submissions! This compares to 2541 for Paris, an 10%+ increase... may the odds be in everyone's favor for what looks to be a record breaking event! (book hotels early!)
events.linuxfoundation.org/kubecon-clou...
loading . . .
KubeCon + CloudNativeCon Europe | LF Events
The Cloud Native Computing Foundationโs flagship conference gathers adopters and technologists from leading open source and cloud native communities.
https://events.linuxfoundation.org/kubecon-cloudnativecon-europe/
1
45
19
I shared the
#Kubeflow
2024 highlights and future roadmap at the latest
#CNCF
WG AI meeting. The Kubeflow community has made incredible progress this year, driving the future of cloud native AI/ML on Kubernetes ๐ Watch the recording:
youtu.be/u4Mf3Jh8v2E?...
View the slides:
bit.ly/4fEql19
loading . . .
Kubeflow Updates 2024
Kubeflow Updates 2024
https://bit.ly/4fEql19
over 1 year ago
0
6
1
This marks a significant milestone for the Kubeflow Community. Stay tuned for details on our upcoming Kubeflow Steering Committee Election 2024!
add a skeleton here at some point
over 1 year ago
0
2
0
If you missed our session announcing the Kubeflow Training V2 at the
#KubeCon
+
#CloudNativeCon
NA, check out the recording. We showcased how weโve made it effortless to fine-tune and train LLMs on Kubernetes. More to come soon!๐
youtu.be/Lgy4ir1AhYw
loading . . .
Democratizing AI Model Training on Kubernetes with Kubeflow Train... Andrey Velichkevich & Yuki Iwai
YouTube video by CNCF [Cloud Native Computing Foundation]
https://youtu.be/Lgy4ir1AhYw
over 1 year ago
0
8
2
Amazing article from the
@cncf.bsky.social
! They asked professional developers about the usefulness of Batch and AI/ML compute technologies. It's great to see that more teams are adopting projects from the
#Kubeflow
ecosystem.
www.cncf.io/reports/cncf...
loading . . .
CNCF Technology Landscape Radar
The Q3 2024 CNCF Technology Landscape Radar examines the adoption and maturity of critical cloud native technologies โ providing detailed insights into multicluster application management solutionsโฆ
https://www.cncf.io/reports/cncf-technology-landscape-radar/
over 1 year ago
0
2
1
I am super excited to present our latest updates on the Kubeflow Training V2 at the
#KubeCon
+
#CloudNativeCon
NA on November 14th ๐ This milestone is the result of an incredible collaboration between the
#Kubernetes
Batch and
#Kubeflow
Training working groups. Don't miss it!
sched.co/1i7nV
loading . . .
KubeCon + CloudNativeCon North America 2024: Democratizing AI Model Training on Kuber...
View more about this event at KubeCon + CloudNativeCon North America 2024
https://sched.co/1i7nV
over 1 year ago
0
0
0
Hi there, let's go back to 2010 ๐
over 1 year ago
0
1
0
you reached the end!!
feeds!
log in