Deep dives on AI infrastructure, MLOps, GPU clusters, and building with agents.
What fits on an RTX 5080 or RTX 5090 Laptop GPU, the memory arithmetic, the open models worth running, Ubuntu 26.04 setup with the NVIDIA stack, Olla…
Evidence instead of proofs, in about 400 lines of Python with real output: slice metrics with bootstrap intervals and a pass/fail contract, Hypothesi…
Install torchcurves, fit and size a curve, choose B-spline or Legendre and the input map, build a monotone calibration curve and an additive model on…
Install cactus-compute, make tool calls, gate on confidence with a fallback to a bigger model, extract structured data, run speech and vision, measur…
Close the AI guidance gap in code: a versioned policy file, an OpenAI-compatible FastAPI gateway, Presidio redaction, per-team tool and MCP allow-lis…
From Lisa Su's 2024 interview to Helios: the 2026 platforms layer by layer, Instinct product lines, memory arithmetic, the CUDA-to-ROCm software map,…
Billing models and idle cost, per-call cost at three volumes, Bedrock in-region vs EU profiles, Converse API with prompt caching and IAM region enfor…