Eight posts on fine-tuning an LLM, written over the back half of 2025 while I was actually doing it. This page puts them in the order you’d want to read them.

The blog lists them newest first, which means you land on deploying to production before you know what fine-tuning is. Not useful. So this is the path instead.

It covers the whole arc: deciding whether you need to fine-tune at all, getting a GPU to cooperate, quantization, the tradeoffs nobody warns you about, evaluating whether the thing actually works, and shipping it.

Every post has an audio version, most have a video, and a few have a two-minute clip if you just want the gist.

Start here

1. Building with LLMs: From Chat to Custom AI When to prompt, when to do RAG, and when fine-tuning is the right answer. Read this before you spend money on a GPU. Sometimes the answer is that you don’t need to fine-tune at all. Video · Audio · Clip

2. What Is LLM Fine-Tuning? What’s actually happening under the hood, explained without the math. Fine-tuning is how a general model becomes your domain expert. Video · Audio

Getting set up

3. So You Want to Fine-Tune a Model (and You Finally Got a GPU!) The setup tips I wish someone had handed me. Hardware, environment, and the things that quietly waste your first weekend. Audio · Clip

4. How to Run Big Models on Small GPUs All about quantization. How to fit a model that doesn’t fit, and what you give up when you do. Video · Audio · Clip

Tuning it

5. LLM Fine-Tuning & Performance Tug-of-War Every knob you turn costs you something somewhere else. This is the tradeoff map. Video · Audio

6. LLM Fine-Tuning Optimization Part 2: Achieving Stability Part two of the tug-of-war. Getting training runs that don’t fall over. Video · Audio

Shipping it

7. Your Fine-Tuned LLM Model Isn’t Ready Yet How to evaluate it before you let anyone near it. Your loss curve looking nice is not evidence that it works. Video · Audio · Clip

8. Opening Night: Launching a Fine-Tuned LLM to Production The part most guides skip. Turning a checkpoint into a service real people can hit, and what happens when the GPU runs out of memory on New Year’s Eve. Video · Audio

Elsewhere

The full set also lives on the blog series page in publication order, and the videos are on YouTube. LinkedIn is the easiest place to reach me if you’re working through this and something doesn’t land.