Back

LLM Post-Training: A Deep Dive into Reasoning LLMs This survey paper provides an in-depth examination of post-training methodologies in Large Language Models (LLMs) focusing on improving reasoning capabilities. While LLMs achieve strong performance

"A Survey on Post-Training of Large Language Models" This paper systematically categorizes post-training into five major paradigms: 1. Fine-Tuning 2. Alignment 3. Reasoning Enhancement 4. Efficiency Optimization 5. Integration & Adaptation 1️⃣ Fin

I really want to work on AI projects but I'm inexperienced with company work, I used to work as a research intern at a lab and was a data science intern at another place. I really want to get into working on hf models, langchain, langgraph, fine-tuni

I fine-tuned 3 models this week to understand why people fail. Used LLaMA-2-7B, Mistral-7B, and Phi-2. Different datasets. Different methods (full tuning vs LoRA vs QLoRA). Here's what I learned that nobody talks about: 1. Data quality > Data quan

AIOps vs LLMOps vs MLOps. I’ve explained each approach in simple steps below. 𝗔𝗜𝗢𝗣𝗦 AIOps applies AI to monitor, detect, and fix problems in IT systems. 1. Decide what issue you want AI to solve (like preventing system crashes). 2. Collect l

My Favorite AI & ML Books That Shaped My Learning Over the years, I’ve read tons of books in AI, ML, and LLMs — but these are the ones that stuck with me the most. Each book on this list taught me something new about building, scaling, and underst

Indian AI launch: Sarvam-M, a 24B open-weights hybrid model built on top of Mistral Small. Sarvam-M achieves a new benchmark across a range of Indian languages, math, and programming tasks, for a model of its size. Here is a detailed technical blog

Researchers at Google DeepMind introduced Semantica, an image-conditioned diffusion model capable of generating images based on the semantics of a conditioning image. The paper explores adapting image generative models to different datasets. Instea

Manual data labeling is one of the biggest bottlenecks in building intelligent image classification systems — especially when the objects are rare, niche, or appear in very few examples. At Gigaversity, we recently faced this challenge while building

Most people learn AI randomly. That’s why they struggle moving from experiments to real production systems later. A strong AI career needs structured depth across fundamentals, systems thinking, modeling, and product execution. Not just model tutor