Sajal Sharma

Introduction The first time I watched an agent write a shell command and run it before I could finish reading it, I was on my own laptop: the one with my API keys, personal notes, calendar, financial records, and years of personal files.

Introduction "The biggest lesson that can be read from 70 years of AI research is that general methods that leverage computation are ultimately the most effective, and by a large margin." — Rich Sutton, The Bitter Lesson (2019) A year ago, I was deep into…

Introduction I've been using Claude Code as a kind of personal assistant for a couple months now.

A reflection on a year of building, teaching, and figuring things out.

Introduction In my previous post, I shared how we built a production-ready risk assessment system using Claude Code, taking it from a lovable.ai prototype to deployed infrastructure.

Introduction I've been using AI coding tools for more than a year now, starting with GitHub Copilot, then jumping to Cursor, from Cursor to Windsurf, and back to Cursor again.

Introduction Note: This post focuses on the conceptual architecture and practical implications of MCP rather than implementation details.

Introduction Welcome to the third installment in my Agentic RAG series!

Introduction Welcome to the second post in my Agentic RAG series!

Introduction The emergence of Large Language Models (LLMs) like OpenAI's GPT-4 and Anthropic's Claude has triggered an explosion of products built on these foundation models.

Introduction A Refresher on RAG Large Language Models (LLMs) or Foundational Models, are powerful but have a fundamental limitation: they can only generate responses based on their training data, which might be outdated or incomplete.

Had the privilege of delivering a guest lecture on Agentic Systems with LangGraph for the Generative AI & Entrepreneurship class (MGT 899) at Yale University on 11 February 2025!

After months of hard work, I’m thrilled to announce my first-ever video course, Building AI Agents with LangGraph, in partnership with O’Reilly!

Introduction Multi-agent systems are all the rage these days, with them being used to improve both capability and performance of AI based workflows.

Introduction What if chunks from a relevant document are not relevant enough for an LLM to answer a question in your RAG system?

Introduction Welcome to this blog post explaining how to deploy StableLM models on AWS Sagemaker Endpoints.

Introduction I recently started the fast.ai course to build up my practical deep learning skills.

Introduction For the day-to-day work of a Machine Learning Engineer or Data Scientist, it is common to use popular ML frameworks like Scikit-learn, Pytorch, etc.