
Blog
Serving, batching, and fine-tuning models in production — written by the engineers who run them.
2 posts · Updated 8 Nov 2026
Latest
Earlier posts
What we cover
Serving, fine-tuning, and life in production.
Inference
Serving and batching at scale, and keeping p99 calm when traffic spikes.
Fine-tuning
LoRA runs, datasets, and what actually moved quality.
Production
Uptime, receipts, and what our incidents taught us.
Run your models here
Serve, batch, and fine-tune on one OpenAI-compatible endpoint.