Log-Loss

Acing your ML interview #1

• ML sys design problems cover real life scenarios — design a video recommendation service, ads ranking system, search query classifier etc.
• The interviewer is looking for both breadth and depth. In breadth the focus is on your understanding of end-to-end ML…

Unified virtual memory for large scale ML workloads

Context:

Continuous testing for ML

Context:

Accelerating multimodal model development

• Typical social network content has a subset of text, videos, audio, image, entities…

Understanding business models for AI companies

We are in the middle of the AI revolution. Apart from AI turbocharging…

A framework for evaluating ml automation

End-to-end ML platforms are the new kids on the block for many enterprises. They are…

Foundation models

We have seen an explosion of ML usage in the last five years across different industry tracks ranging from ads ranking, search…

Drawbacks of using AUC-ROC as measure

Context :

Testing ML Components

Introduction to ml component tests

What is a ML component ?

• A machine learning component is an individual API or a group of APIs and utilities that implement a particular task in the model lifecycle.