Using AI to Port a Wi-Fi Driver from Linux to FreeBSD: A Case Study
Direct AI Code Porting Attempt
The developer had a 2016 MacBook Pro with a Broadcom BCM4350 Wi-Fi chip that lacks native FreeBSD support. The typical workaround is wifibox - a Linux VM with PCI pass-through. Instead, they attempted to port the Linux brcmfmac driver (ISC license) directly to FreeBSD using AI.
They cloned the brcmfmac subtree and asked Claude Code to make it work for FreeBSD, specifically pointing to the iwlwifi driver as a reference for using LinuxKPI (FreeBSD's Linux kernel compatibility layer). The initial attempt produced a module that compiled but didn't function because the test VM lacked the actual hardware.
When testing with the actual PCI device, the kernel panicked. Claude Code attempted to fix panics by adding #ifdef __FreeBSD__ wrappers and building FreeBSD-specific shims and callbacks, but warned the project was becoming "very complicated and messy." The resulting diff was significantly larger than expected, and the driver remained non-functional.
Specification-First Approach
Inspired by Armin Ronacher's experience with Claude Opus and Pi agent, the developer switched approaches. Recognizing their task was narrow (one chip, only PCI, only Wi-Fi client), they spawned a fresh Pi session and asked the agent to write a detailed specification of how the brcmfmac driver works with focus on BCM4350.
They explicitly set the audience as "readers tasked with implementing the specification in a clean-room environment" and asked for explanations "to the bits." The agent produced an 11-chapter specification covering:
- 00-overview.md
- 01-data-structures.md
- 02-bus-layer.md
- 03-protocol-layer.md
- 04-firmware-interface.md
- 05-event-handling.md
- 06-cfg80211-operations.md
- 07-initialization.md
- 08-data-path.md
- 09-firmware-commands.md
- 10-structures-reference.md
The developer notes that AI-generated specifications require verification, as one "can't just trust what AI has written."
📖 Read the full source: HN AI Agents
👀 See Also
How to Use Claude Code Effectively: A Developer's Experience Building a Full SaaS App
A developer with SaaS experience since 2021 built a complete spaced repetition app called codefluent.app using Claude Code, emphasizing that success depends on writing detailed technical specifications rather than vague prompts. The project used SvelteKit, PostgreSQL with Drizzle ORM, Better Auth, OpenRouter, Stripe, CodeMirror 6, Tailwind v4, and Railway.
DevOps engineer uses Claude Code to build custom terminal app
A DevOps/SRE engineer with years of experience used Claude Code to build a terminal application they had envisioned but couldn't complete alone. The AI handled scaffolding and integrations while the engineer focused on product decisions.
Claude Mobile Workflow: Brainstorm Features on Phone, Get Autonomous Implementation
A developer shares a workflow where they brainstorm features and bug fixes with Claude on their phone while mobile, then have a daemon script automatically implement well-defined tasks by creating Linear issues and spinning up Claude Code agents to handle implementation, testing, and deployment to staging.
Running a 6-agent behavioral coaching pipeline on self-hosted Qwen3 235B with vLLM
A developer built a 6-agent cognitive pipeline for behavioral coaching running entirely on self-hosted Qwen3 via vLLM, using Qwen3 30B on 2x RTX 4090s for development and Qwen3 235B on RunPod A40 pods for production.