Shipping custom models at scale from fine-tuning to inference | BRK234

Name: Shipping custom models at scale from fine-tuning to inference | BRK234
Uploaded: 2026-06-04T12:30:50.000Z
Channel: Microsoft Developer

Machine learning Fine Tuning Model serving Inference optimization AI infrastructure Production ML Model deployment MLOps Microsoft Build Scaling AI models

Microsoft Developer June 4, 2026

AI summary

A panel of practitioners from Microsoft, Unsloth, and stealth startups discusses end-to-end ML model deployment, from fine-tuning strategies to production serving. The conversation covers infrastructure decisions for model serving, techniques for optimizing inference cost and latency, and real-world approaches to taking custom models to production at scale. Ideal for ML engineers and developers building AI-powered products who want practical insights on shipping models efficiently.