Mar 24 · 4 min read · What do AI engineers hate most? Not hyperparameter tuning. Not waiting for GPUs. It's setting up a distributed training job on Kubernetes. PyTorchJob, TFJob, MPIJob, XGBoostJob, PaddleJob, JAXJob. Six CRDs, six YAML formats, six knowledge domains. Sw...
Join discussionMar 24 · 9 min read · PyTorch has emerged as an important device of machine learning and artificial intelligence. It is what businesses are currently using to create intelligent platforms that process information and make
Join discussion
Mar 22 · 4 min read · Local LLM training has a dirty secret. Everyone talks about the magic of custom weights, but nobody talks about the grueling reality of babysitting PyTorch scripts. You set up your data, configure you
Join discussion
Mar 11 · 8 min read · Part 1 of an ongoing series documenting my hands-on journey into PyTorch and deep learning as a software engineer transitioning into AI Engineering. Why I'm Writing This I'm a full-stack software en
Join discussion
Mar 9 · 9 min read · HAR Series — Part 4 · ~12 min read A quick note: If you noticed a bigger gap than usual between posts, my apologies! Things have been hectic on personal and professional fronts — in the best way pos
Join discussion
Mar 6 · 13 min read · I highly recommend to have a knowledge of machine learning models or atleast the basics The Core Idea: Transformers Before transformers, the industry relied on RNNs and LSTMs. The paper "Attention
Join discussionFeb 17 · 6 min read · Artificial Intelligence isn't just "the future" anymore—it is the present. From recommendation engines on Netflix to fraud detection in banking, Machine Learning (ML) engineers are building the infrastructure of the modern world. But with thousands o...
Join discussionFeb 16 · 10 min read · Every once in a while, you come across these concepts that are so tricky to grasp, they tickle the insides of your brain. Introducing RoPE, he gave me a hard time last week. Hi future-me, who forgot how RoPE works, this is for you. RoPE stands for Ro...
Join discussion
Feb 16 · 3 min read · Introduction When I started learning about LLMs, terms like self-attention and Transformers felt almost magical. So instead of only using libraries, I decided to build a language model step by step — starting from a very simple baseline and ending wi...
Join discussion
