engblogs

This blog post explores the use of Contextual Bandit (CB) models in large action spaces at Instacart, discussing the challenges, potential solutions, and the application of CB models to ranking and retrieval problems in e-commerce.

Introducing AvroTensorDataset, a high-performance TensorFlow dataset for efficiently processing Avro data in machine learning models

This blogpost is about experimenting at scale and optimizing the Spotify Home experience for enhanced personalization and user value.

A post by Peter Davio, CTO at Black Swan Data, discussing strategies for handling petabyte-level learning at Snorkel AI's The Future of Data-Centric AI Summit in 2022.

This blog post discusses the on-device acceleration of large diffusion models via GPU-aware optimizations, focusing on techniques such as enhanced attention modules for memory efficiency, Winograd fast convolution for 3x3 convolution layers, and specialized operator fusion for memory efficiency.

Exploring the diversity and nuances of Spanish dialects across the globe.

Creating a multilingual automatic translation pipeline using Amazon Translate Active Custom Translation

Empowering digital transformation with Google Cloud's innovative AI and machine learning tools across Document AI, wind energy optimization, and AlphaFold on Vertex AI.

Recognition of David Autor's research on the effects of technological change and globalization on jobs and earnings prospects for workers.

Exploring Grammarly's efforts in achieving superhuman communication assistance

This blogpost is about safely migrating Netflix's mobile apps to GraphQL, covering strategies such as AB Testing, Replay Testing, and Sticky Canaries.

The blog post describes the work done to create indoor views in Immersive View using neural radiance fields (NeRF) for realistic, multi-dimensional reconstructions of spaces, including the pipeline for capturing photos, image processing, and scene reproduction, and the use of interactive 360-degree videos for a scalable user experience.