Assaf PinhasiinFeature Stores for MLFeature pipelines and feature stores — deep dive into system engineering and analytical tradeoffsIntroduction17 min read·Dec 8, 2022--1--1
Assaf PinhasiWhat I’ve learnt about deep work after years of being distractedWhat is deep work?9 min read·Nov 23, 2022--1--1
Assaf PinhasiinTowards Data ScienceFrom raw videos to GAN training — implementing a data pipeline and a lightweight Deep Learning…Introduction15 min read·Oct 20, 2022----
Assaf PinhasiinTowards Data ScienceDeep Lake — an architectural blueprint for managing Deep Learning data at scale — part IIntroduction15 min read·Jun 10, 2022----
Assaf PinhasiWhat I’ve learnt from two years of being an expert hands-on technology consultantWhat led me here?14 min read·Dec 26, 2021--5--5
Assaf PinhasiinTowards Data ScienceHow to run CPU-based Workloads for Deep Learning Using Thousands Of Spot Instances on AWS and GCP…Deep learning is notorious for consuming large amounts of GPU resources during training. However, there are multiple parts within the Deep…10 min read·Jul 26, 2021----
Assaf PinhasiinPyTorchA Step by Step Guide to Building A Distributed, Spot-based Training Platform on AWS Using…This is part II of a two-part series, describing our solution for running distributed training on spot instances using TorchElastic and…10 min read·Jun 29, 2021----
Assaf PinhasiinPyTorchHow 3DFY.AI Built a Multi-Cloud, Distributed Training Platform Over Spot Instances with…Deep Learning development is becoming more and more about minimizing the time from idea to trained model.9 min read·Jun 17, 2021--3--3
Assaf PinhasiGreat writeup, thanks for sharing.I am under the impression that the backwards pass is in fact the synchronization barrier.1 min read·May 26, 2021--1--1