InData Science CollectivebyGary A. Stafford·15h agoAutomating Model Training and Delivery from Amazon SageMaker AI to Amazon EKSAn end-to-end walkthrough of model tuning, evaluation, explainability, registration, and approval-driven Kubernetes deployment.
Gary A. Stafford·6d agoDeploying and Managing Agents on Amazon Bedrock AgentCore Built with OpenAI Agents SDKHow AgentCore Runtime, Identity, Memory, OpenAI Agents SDK, and OpenAI GPT-5.6 models work together to deploy, host, and observe agents
InData Science CollectivebyGary A. Stafford·Jul 4Building Agentic Systems with the OpenAI Agents SDK on Amazon Bedrock MantleBuilding agents with the OpenAI Agents SDK, incrementally, across six increasingly feature-rich versions
Gary A. Stafford·Jul 1Using OpenAI Codex CLI with LiteLLM AI Gateway and Amazon Bedrock MantleUsing LiteLLM as a gateway between Codex CLI and OpenAI models available on Amazon Bedrock MantleA response icon1A response icon1
InData Science CollectivebyGary A. Stafford·Jun 29Running LiteLLM as a Proxy in Front of Multiple Model ProvidersA practical guide to consolidating LLM provider access for individual developers and platform teams.A response icon3A response icon3
Gary A. Stafford·May 18The Introspection Illusion in Large Language ModelsWhy confabulated metacognition is a distinct subtype of hallucination, and why it matters.
Gary A. Stafford·Apr 22Building a Real-Time Voice AI Stack on AWS with NVIDIA NemotronDeploying a voice-to-voice app on Amazon EKS, using open-weight NVIDIA Nemotron ASR, Qwen, and Magpie TTS
InArtificial Intelligence in Plain EnglishbyGary A. Stafford·Apr 11Shipping Real‑Time Voice AI on AWS with Kyutai Unmute and Gemma 4Real-time, sub-second voice conversations powered by open-source AI, deployed, secured, and scaled on Amazon EKS.
InData Science CollectivebyGary A. Stafford·Mar 31High-quality, Low-cost Image Generation and Editing with QwenBuild and deploy a Qwen Image and Qwen Image Edit app for text-to-image generation and image-to-image editing on your GPU-based workstation…
InData Science CollectivebyGary A. Stafford·Mar 2High‑Quality Long‑Form TTS with Qwen3 Open‑Weight ModelsA practical guide to building robust long‑form text-to-speech pipelines with open models on consumer-grade hardware.A response icon1A response icon1
