https://blog.bjdean.id.au/tags/azure/Recent content in Azure on Brad's BlogHugo -- 0.152.2en-auBradley DeanThu, 20 Nov 2025 17:30:51 +1100https://blog.bjdean.id.au/2025/11/fine-tuning-azure-ai-foundry-practical-lessons/Thu, 20 Nov 2025 17:30:51 +1100https://blog.bjdean.id.au/2025/11/fine-tuning-azure-ai-foundry-practical-lessons/<h2 id="tldr">TL;DR</h2> <p>Azure AI Foundry makes fine-tuning accessible with straightforward data preparation requirements, but costs can accumulate quickly during experimentation. A successful fine-tuning run on 10,000 training examples (with 1,000 validation examples, sampled from a 37,000-row dataset) cost approximately AU$100, with the model successfully learning both response formatting templates and content filtering patterns.</p> <p>Lessons learned: understand the model and training technique settings and data requirements, monitor your training metrics early to avoid over-training and watch your spending closely since billing data lags by 24-48 hours.</p>