https://michalwojdylak.com/blog Building production AI systems, LLM infrastructure, inference platforms and cloud-native ML solutions. en Wed, 10 Jun 2026 00:00:00 GMT https://michalwojdylak.com/blog/first-post https://michalwojdylak.com/blog/first-post A practical walkthrough of the infrastructure decisions behind serving large language models reliably — from GPU selection to batching, autoscaling, and observability. Wed, 10 Jun 2026 00:00:00 GMT LLM Infrastructure Inference https://michalwojdylak.com/blog/optimizing-gpu-inference-costs https://michalwojdylak.com/blog/optimizing-gpu-inference-costs Quantization, batching, and right-sizing strategies that reduced our inference bill by 60% while keeping p99 latency flat. Fri, 22 May 2026 00:00:00 GMT Inference Optimization Cost https://michalwojdylak.com/blog/building-mlops-platform-kubernetes https://michalwojdylak.com/blog/building-mlops-platform-kubernetes The core building blocks of a production MLOps platform — model registry, CI/CD for models, and safe rollouts with canaries and shadow deployments. Thu, 30 Apr 2026 00:00:00 GMT MLOps Kubernetes Infrastructure