Aguasto / learn
The engineering behind AI.
Practical guides to memory, serving architecture and reliable operations. Explanations grounded in official documentation, with assumptions you can inspect.
Selected publications
Model weights are only the beginning of the memory budget
Help engineers separate a transparent weight-storage estimate from the workload-dependent memory required to serve a model.
Prefill, decode and the cost of separating them
Explain the two serving phases and when to evaluate disaggregation without claiming automatic throughput gains.
Choosing between vLLM, SGLang and TensorRT-LLM
Compare documented serving-engine roles and build a selection procedure rather than an unsupported performance ranking.
From Power to Inference—Building a Distributed AI Compute Network
Connect physical capacity, server/runtime assumptions and completed inference work through a transparent conceptual measurement methodology.
Distributed inference reliability begins with the failure boundary
Distinguish replica recovery, control-plane recovery and regional failover, and specify the evidence needed before publishing availability claims.