There's currently 40841658 ways of doing ML

It takes time for a Canonical Stack to develop in any new field. Before it develops and unleashes a tidal wave of new innovation, it’s often super hard for anyone to see where it’s all going.

  §

Machine Learning Lifecycle

  • Highly Developed
  • Moderately Developed
  • Still Developing
  • Lightly Developed
  • 0. Resource Management, DL Enablement
  • 1. Data Gathering, Transformation
  • 2. Experimenting, Training, Tuning, Testing
  • 3. Productionization, Deployment, Inference
  • 4. Monitoring, Auditing, Management, Retraining
0.1.2.3.4.
Acceldata
Airbytes
Airflow
Algorithmia
Argo
Arize AI
Arthur
AtScale AI-Link
Auger.AI
Azkaban
Bodywork
ClearML
cnvrg.io
Coiled
Comet
DAGsHub
Dagster
data.ai
Delta Lake
Determined AI
DoltHub
Dud
Fiddler
Flyte
Git LFS
Gretel.ai
Hopsworks
Iguazio
Inferrd
Iterative
ixpantia
Kedro
Keepsake
Kubeflow
lakeFS
LatticeFlow
Luigi
Maiot
Marquez
Metaflow
MindsDB
MLFlow
ModelOp
Molecula
Neptune
Neu.ro Inc.
New Relic
NimbleBox.ai
OctoML
Orchest
Pachyderm
Picsellia
Pipeline.ai
Prefect
Prevision.io
PrimeHub
Qlik
Quilt
Sama
Seldon
Sisense
Snitch AI
Spell
Superb AI
Superwise.ai
Tecton
Tekton
Terminus DB
TIBCO
Toloka
TruEra
UbiOps
Valohai
Weights & Biases
WhyLabs
YData
  §

End-to-end ML platforms typically take 15–36 months to catch up on feature parity, but the products will launch much sooner than that to create thought leadership and get product feedback.

  §
  §