https://rwatsh.github.io/A minimal, responsive and feature-rich Jekyll theme for technical writing. 2025-10-27T08:48:44-05:00 Watsh Rajneesh https://rwatsh.github.io/ Jekyll © 2025 Watsh Rajneesh /assets/img/favicons/favicon.ico /assets/img/favicons/favicon-96x96.png 2025-10-27T08:41:00-05:00 2025-10-27T08:41:00-05:00 https://rwatsh.github.io/posts/model-context-protocol/ rwatsh

Model Context Protocol (MCP) https://modelcontextprotocol.io/ Below are notes from: https://learn.deeplearning.ai/courses/mcp-build-rich-context-ai-apps-with-anthropic What is MCP? MCP = Standardizes how LLM apps interact with external systems.   Stateless connection is useful for cases where we spin up the MCP server on-the-fly (server less compute). If the server is not deployed as ephem...

2025-10-20T08:15:00-05:00 2025-10-20T08:15:00-05:00 https://rwatsh.github.io/posts/langgraph-agent-with-tool-calling/ rwatsh

LangGraph Agent with Tool Calling Example This example demonstrates a LangGraph agent that can intelligently decide when to call external tools during its reasoning process. The agent uses Groq to access meta Llama3 LLM and three tools: Google Knowledge Graph Search, Organic Google Search, and Firecrawl for web scraping. Code is given below: import os from typing import Literal from dotenv i...

2025-10-20T01:50:00-05:00 2025-10-20T01:50:00-05:00 https://rwatsh.github.io/posts/langgraph-agent-with-memory/ rwatsh

Example of a LangGraph agent with memory that allows iterative refinement of LLM responses based on user feedback. The agent uses Ollama LLM and LangGraph’s InMemorySaver for state persistence across interactions. from typing import Annotated from typing_extensions import TypedDict from langgraph.graph import StateGraph, START, END from langgraph.graph.message import add_messages from langgrap...

2025-10-20T00:09:00-05:00 2025-10-20T01:00:43-05:00 https://rwatsh.github.io/posts/basic-langgraph-agent/ rwatsh

LangGraph Basic Agent Example LangGraph is a new orchestration framework from LangChain for building, managing, and deploying long-running, stateful agents. It is part of the LangChain stack which includes: Orchestration - LangGraph Integrations - LangChain Evals + Observability - LangSmith Deployment - LangGraph Platform Each stack component/product can be used independently of ot...

2025-10-19T10:41:00-05:00 2025-10-20T00:29:54-05:00 https://rwatsh.github.io/posts/langchain-with-ollama-cloud-models/ rwatsh

Ollama provides a way to run large language models (LLMs) locally or in the cloud. It supports various models including open-source ones. Langchain, a popular framework for building applications with LLMs, can be integrated with Ollama to leverage these models. Below is a simple example of how to use Langchain with Ollama cloud models. Pre-requisites: Install Ollama and set up your cloud i...