Advait Shinde·Aug 4VLMs: Paradigms and Trade-offsLLMs live in a tidy universe. Every input gets sliced into discrete tokens, each one a pointer into a finite vocabulary matrix ………
Advait Shinde·Mar 25CNNs : breakdown and breakthroughs from 1998 to 2019 !Convolutional neural networks got Deep Learning on the board. Before CNNs, deep learning was mainly comprised of Fully Connected Layers!
Advait Shinde·Mar 20How Reinforcement Learning quietly runs your Recommender Systems!An article about use of multi-armed and contextual banditsthe algorithms that Netflix, Spotify, Stitch Fix, and DoorDash use to stop…
Advait Shinde·Jan 25LLM Scaling laws and their relevance in 2026!We have all heard about Open AI / Kaplan Scaling laws and Chincilla scaling laws, but these were earlier in the decade, do the current…A response icon2A response icon2
Advait Shinde·Oct 28, 2025Finding the Perfect Replacement: How ML Recommender Systems can aid Valorant Transfers!A deep dive into player performance analytics and recommendation systems for the 2025 transfer season
Advait Shinde·Apr 14, 2025Unpacking the Google ADK: What Developers Need to KnowGoogle just launched their open source agent development framework, and here are my takeaways, if interested, give this blog a read :DA response icon1A response icon1
Advait Shinde·Apr 7, 2025Behemouth, Maverick and Scout!Meet the Meta Llama 4 models with a ton of features and impressive benchmark results!
Advait Shinde·Mar 17, 2025Revolutionizing AI Agents: LangMem Long Term MemoryImagine an AI assistant that not only remembers your current conversation but also recalls your preferences, past experiences, and even…
Advait Shinde·Jan 4, 2025RAG without LangChain and LlamaIndex?Large Language Models are a subset of artificial intelligence that utilize deep learning techniques, particularly transformer…
Advait Shinde·Nov 5, 2024Understanding Flow Engineering: A New Paradigm in AI Development via LaIn the ever-evolving landscape of artificial intelligence, a new concept has emerged that promises to reshape how we approach software…