~/talks

David Aronchick

what they talk about

all 10 in the filters

where they keep going back

all 14 in the filters

on a stage with

all 5 in the filters

who talks about the same things

titleconferencedate
THE SANITY CHECK: OBSERVABILITY FOR AI DATA PIPELINESAI SRE Summit 20262026-05-12
A chatbot answers “365 days” but the true policy is 30 days because a rogue Salesforce feed pollutes embeddings; the pipeline is a black box so top-line…
Talk on
The Sanity Check: Observability for AI Data PipelinesCloud Native AI + Kubeflow Day 20262026-03-23
Claim: attach provenance at ingestion so a bad LLM answer can be traced to the exact source. Shows an open spec called data bill of materials and an OSS…
Talk on
Taming the Data TsunamiPyData Seattle 20252025-11-07
Claims a concrete open‑source playbook that shifts filtering and governance upstream and policy-as-code for distributed pipelines that reduce dataset size by…
Talk on
Universal AI: Execute Your Models Where Your Data (And Users) AreOpen Source Summit North America 20252025-06-23
Claim: run ML next to users and data so pipelines stay fast, cheap and compliant. Platform Bakalao (Expanso) runs on WASM, Docker, Wom and arbitrary binaries…
Talk on
The MODERN Modern Data Stack: Building an Open Distributed Data Warehouse Beyond...Open Source Summit North America 20252025-06-23
Argues that bringing compute to data cuts egress, latency and compliance risk and shows a working stack: deploys Bacalhau agents on VMs across AWS GCP and…
Talk on
Kubernetes Cross-Zone/Region Simplified: Harnessing Bacalhau for Efficient Distributed ComputeKubernetes on Edge Day Europe 20252025-04-01
Argues that moving compute to data cuts transfer costs and compliance risk, demonstrates an open source platform Bacalhau wired into the Kubernetes CRD so one…
Talk on
WASM in Production: Building the Missing Infrastructure Layer with BacalhauWasm I/O 20252025-03-27
Claims that Bacalhau can deploy the same WASM job across clouds edge and local with zero code or job changes, demonstrates a live run of ~30 simulated sensors…
Talk on
Universal and Reproducible ComputeIPFS Camp 20242024-07-11
Bacalhau implements Compute over Data by running containers, WASM, or arbitrary binaries where the data lives, assigning a CID per job and per input, and…
Talk on
Revolutionizing the Big Data Age With Compute over DataPyData Global 20222022-12-01
Claims a usable project called Bacalhau that runs Docker/WASM jobs where the data lives, cutting the need to move 100 terabytes across networks. Demonstrates…
Talk on
The SAME Project: A Cloud Native Approach to Reproducible Machine LearningKubernetes AI Day Europe 20212021-05-04
SAME provides a one-line install and a CLI that bootstraps a local k3s cluster, deploys Kubeflow, and runs pipelines described in a declarative same file; it…
Talk on
Owned by Statistics: Using Kubeflow to Defend vs Attacks on Your ML ModelsKubeCon + CloudNativeCon North America 2020 Virtual2020-11-17
Shows how a cloud-native MLops pipeline built with Kubeflow Pipelines plus Kale (notebook→pipeline), Katib for hyperparameter search, Rok snapshotting and ML…
Talk on
Owned By Statistics: How Kubeflow & MLOps Can Help Secure...Scale By the Bay 20202020-11-12
Shows concrete attacks and defenses: builds a toy circle-vs-square failure and links it to real failures like a snow detector for huskies, stickers that flip…
Talk on
KubeFlow BoFKubeCon + CloudNativeCon Europe 20192019-05-20
Room discussion focuses on actionable work: a KFServing standard API is being defined to let one model package deploy to TensorFlow Serving, Seldon, or cloud…
Talk on
Towards Kubeflow 1.0, Bringing a Cloud Native Platform For ML to KubernetesKubeCon + CloudNativeCon Europe 20192019-05-20
Project claims a cloud-native ML platform on Kubernetes that already runs core pieces in production: Kubeflow pipelines for orchestration, Katib for…
Talk on
Managing Machine Learning in Production with Kubeflow and DevOpsKubeCon + CloudNativeCon Europe 20192019-05-20
Live demo builds three container images via Azure Pipelines on every merge, triggers a Kubeflow pipeline that runs MobileNet transfer learning on GPUs…
Talk on
A Year of Democratizing ML With Kubernetes & KubeflowKubeCon + CloudNativeCon China 20182018-11-13
Project hits 0.3 with tooling that actually works: KFctl one-line install and a GCP click-to-deploy that provisions clusters and apps, a Pipelines Python SDK…
Talk on
Machine Learning on Kubernetes Birds of a FeatherKubeCon + CloudNativeCon China 20182018-11-13
Roadmap and demos focus on making Kubeflow production ready with a stable API toward 1.0, plus a newly launched Kubeflow Pipelines workflow orchestrator that…
Talk on
Kubeflow Deep DiveKubeCon + CloudNativeCon Europe 20182018-05-02
Kubeflow presents a Kubernetes-native ML platform built from ksonnet packages and custom resources that already supports TF operator, a forthcoming PyTorch…
Talk on
Keynote: Kubeflow ML on KubernetesKubeCon + CloudNativeCon Europe 20182018-05-02
Announces Kubeflow 0.1/1.0 as a packaged, Kubernetes-native ML stack with Jupyter, TensorFlow training, TensorBoard and model serving; demo shows a local…
Talk on
Conquering a Kubeflow Kubernetes Cluster with ksonnet, Ark, & SonobuoyKubeCon + CloudNativeCon Europe 20182018-05-02
Demo shows Kubeflow deployed with ksonnet, validated with Sonobuoy (conformance run 30–60 minutes that surfaces multiple failing e2e tests), and migrated with…
Talk on
''Hot Dogs or Not" - At Scale with Kubernetes [I]KubeCon + CloudNativeCon North America 20172017-12-06
Portable ML pipeline that runs unchanged from laptop minikube to GKE with GPUs, packaged as cube flow components deployable via kubectl. Includes JupyterHub, a…
Talk on
How Google Cloud Hosts and Manages Kubernetes at ScaleCloudNativeCon + KubeCon Europe 20172017-03-29
Google Container Engine claims to run Kubernetes at true production scale — spinning up many thousands of clusters in ~3–5 minutes via a single API/button by…
Talk on
Keynote: Kubernetes - 1.4 and BeyondKubeCon + CloudNativeCon North America 20162016-11-08
Kubernetes 1.4 ships with a new installer and workflow: kubeadm packages components and boots a master then nodes join with a token; networking adds a…
Talk on
~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.