ML infrastructure · Inference · Distributed systems
Engineering systems that make AI work at scale.
Practical architecture deep dives, engineering trade-offs, and lessons from building production machine learning systems. Written by Lokesh Jain.
What I write about
01 / Inference & serving
Model serving, GPU workloads, batching, latency, throughput, and multimodal AI systems.
02 / Distributed systems
Scheduling, orchestration, resource management, reliability, and scalable execution.
03 / ML platforms & LLMOps
Training infrastructure, active learning, experiment tracking, deployment, and production workflows.