Tarush Singh·May 5Stop Letting LLMs Struggle with Tables — Make Them Think FirstMost people try to use LLMs with tabular data in the most direct way possible:
Tarush Singh·Mar 19Stop Sending Entire Chunks to Your LLMUsing Context Extraction to Reduce Tokens and Speed Up RAG Pipelines Retrieval-Augmented Generation (RAG) has become the default…
Tarush Singh·Dec 23, 2025Designing AI Systems That Get Cheaper Over TimeMost production GenAI systems scale cost linearly: more users → more LLM calls → higher bills.
Tarush Singh·Dec 16, 2025Tracking LLM Hallucinations Using ReAct-Style Reasoning (Without Telling the Model the Rules)Most evaluations of Large Language Models (LLMs) focus on a single question:
Tarush Singh·Oct 20, 2025🧠 Building a Self-Improving SQL Agent with Milvus and Reinforcement-Inspired MemoryTurning a static prompt into an evolving reasoning system
Tarush Singh·Jul 28, 2025From Goldfish to Elephant: How Vector Database Memory Makes AI Agents Actually SmartWhy Does My AI Agent Forget Everything?
Tarush Singh·Jun 26, 2025MCP from Scratch: Building a Python Agent that Understands and Executes Your CodeA hands-on guide to building an AI agent powered by docstrings, LangChain, and the Model Context Protocol (MCP)A response icon2A response icon2
Tarush Singh·May 29, 2025LLM Context enhancementLarge Language Models (LLMs) like GPT-4 and Claude are impressive. They can generate poems, write code, summarize articles, and answer…A response icon1A response icon1
Tarush Singh·Apr 19, 2025🚀 Supercharging PyTorch Training: 10 GPU Optimizations with Functional CodeWe have all been thereA response icon1A response icon1