Breexzed

Infrastructure engineering · Linux, Go, from the kernel up

Building An HTTP Server From Scratch

2026-09-13 · project go http learning systems

Building An HTTP Server From Scratch

I attempted building an HTTP server once before, as part of the Backend from First Principles: Complete Series track. I got stuck shortly after establishing a TCP connection and sending requests to it — hardcoded request, hardcoded response. Those early stages live in that project's git commit logs.

The first attempt lives in httpd-go on GitHub.

Later I came across the insights from On Learning, Finding Projects, and Building Towards a Mind of Gold, and I have a much clearer picture now of how I want to approach building this. I'm still unclear on a lot of the implementation details, but I plan to use this exact medium — this log — to break through them. I know I can do it. I will overcome every block.

## Project Overview

I read an intern's article on rate limiting at incident.io, as part of the protocols from On Learning, Finding Projects, and Building Towards a Mind of Gold. Here's the blog: Rate limiting for beginners.

I took it to ChatGPT and used some of the prompts from the protocol I'd followed. The infographic above is the result of that session.

The why, what, and how (task, tool, scope)

I'd already been attempting to build an HTTP server. Then I found the rate-limiting blog, and I wanted to build one at a scale I could actually manage. So I decided — why not use the HTTP server as the base and scale toward the rate limiter, since they sit in the same field? ChatGPT helped me concretize that instinct and gave me the roadmap (and the infographic above) to build:

A small HTTP server in Go that starts as a bare process accepting requests, and gradually evolves into a measurable, resilient, concurrent, performance-tested piece of infrastructure.

Repo: httpd-go

Stage Move Capability
1 BUILD HTTP server
2 OBSERVE Logging
3 REMEMBER Request counter
4 CONTROL Rate limiter
5 MODEL ENTITIES Per-client state
6 COMPOSE Concurrent clients
7 HANDLE CONCURRENCY Timeouts
8 BOUND TIME Connection limits
9 CONTROL LIFECYCLE Graceful shutdown
10 MEASURE Metrics
11 STRESS Load testing
12 BREAK Failure injection
13 COMPARE Benchmarking
14 DIAGNOSE Profiling
15 DISTRIBUTE Distributed rate limiting

This folder is where I document the process of my learning — the struggles, the concepts I come across, the insights, and what changes in me as I keep crossing the boundary of my own understanding. Using documentation, experiments, deliberate information collection, and agents primarily as research, debugging, and critique instruments.

I wish myself good luck.

## Stages