<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
  <title>Parth Sharma</title>
  <subtitle>Building agents, tooling, and writing about it. Go Agent SDK, agentic frameworks, and systems engineering.</subtitle>
  <link href="https://parthshr370.github.io/feed.xml" rel="self" type="application/atom+xml"/>
  <link href="https://parthshr370.github.io" rel="alternate" type="text/html"/>
  <updated>2026-05-05T08:36:06+00:00</updated>
  <id>https://parthshr370.github.io/</id>
  <author>
    <name>Parth Sharma</name>
  </author>
  
  
  <entry>
    <title>Let&apos;s learn LLM streaming from first principles</title>
    <link href="https://parthshr370.github.io/blogs/let-me-teach-you-llm-streaming-from-first-principles/" rel="alternate" type="text/html"/>
    <id>https://parthshr370.github.io/blogs/let-me-teach-you-llm-streaming-from-first-principles/</id>
    <summary>Understanding LLM streaming from first principles by building it into a Go Agent SDK. Covers SSE, stream events, provider parsers, accumulation, tool calls, and RunStream.</summary>
    <author>
      <name>Parth Sharma</name>
    </author>
  </entry>
  
  <entry>
    <title>Anatomy of Subagents</title>
    <link href="https://parthshr370.github.io/blogs/anatomy-of-subagents/" rel="alternate" type="text/html"/>
    <id>https://parthshr370.github.io/blogs/anatomy-of-subagents/</id>
    <summary>How nested agent runs help you delegate, isolate context, and build cleaner multi-agent systems.</summary>
    <author>
      <name>Parth Sharma</name>
    </author>
  </entry>
  
  <entry>
    <title>Pairing your AI agent with 600+ MCP tools seamlessly</title>
    <link href="https://parthshr370.github.io/blogs/pairing-agents-with-mcp/" rel="alternate" type="text/html"/>
    <id>https://parthshr370.github.io/blogs/pairing-agents-with-mcp/</id>
    <summary>How the Model Context Protocol lets your agent connect to 600+ tools without writing custom integrations for each one.</summary>
    <author>
      <name>Parth Sharma</name>
    </author>
  </entry>
  
  <entry>
    <title>JSON, Marshals and Go</title>
    <link href="https://parthshr370.github.io/blogs/json-marshals-and-go/" rel="alternate" type="text/html"/>
    <id>https://parthshr370.github.io/blogs/json-marshals-and-go/</id>
    <summary>A deep dive into how Go handles JSON data structures using Marshalling and Structs. The foundation for understanding how agent SDKs talk to LLM APIs.</summary>
    <author>
      <name>Parth Sharma</name>
    </author>
  </entry>
  
  <entry>
    <title>How I built my own Langchain from scratch in Go (Agent SDK)</title>
    <link href="https://parthshr370.github.io/blogs/how-i-built-my-own-langchain-from-scratch-in-go/" rel="alternate" type="text/html"/>
    <id>https://parthshr370.github.io/blogs/how-i-built-my-own-langchain-from-scratch-in-go/</id>
    <summary>Building an agent SDK from first principles in Go. Covers API structs, message builders, history as an append-only array, and how marshalling bridges Go types and JSON.</summary>
    <author>
      <name>Parth Sharma</name>
    </author>
  </entry>
  
  <entry>
    <title>How I built a tool calling engine for my Go agent SDK</title>
    <link href="https://parthshr370.github.io/blogs/how-i-built-tool-calling-engine-for-my-agent-sdk/" rel="alternate" type="text/html"/>
    <id>https://parthshr370.github.io/blogs/how-i-built-tool-calling-engine-for-my-agent-sdk/</id>
    <summary>Building the tool calling engine from first principles in Go. Covers registration, schema generation via reflection, execution, and how tool results flow back to the LLM.</summary>
    <author>
      <name>Parth Sharma</name>
    </author>
  </entry>
  
</feed>
