Tech Meridian
← ENTITY INDEX
TOPIC · ENTITY #209
Related event timeline, sources and context from the news index.
AWS Machine Learning published a walkthrough that demonstrates customizing Qwen3-8B using supervised fine-tuning (SFT) and reinforcement learning with verifiable rewards (RLVR) via Amazon SageMaker serverless model customization, then deploying it for asynchronous inference to create a cost-efficient product tagging system.
AWS Machine Learning describes how Abnormal AI deployed Amazon Bedrock AgentCore’s Code Interpreter as an ephemeral compute "scratch pad" for the agents powering its real‑time email threat detection at billion‑message scale, and outlines sandbox design decisions and practical lessons for running Code Interpreter in production.
Ninth Wave developed Compass, a multi-agent AI onboarding assistant running on Amazon Bedrock AgentCore that validates bank APIs against Financial Data Exchange (FDX) standards, assigns compliance scores, and reduces open-finance onboarding from weeks to minutes while meeting SOC 2 and PCI DSS requirements, according to AWS Machine Learning.
AWS Machine Learning published an article offering an 8-step decision framework to help pick generative AI customization approaches on AWS, covering prompt engineering, RAG, fine-tuning, continued pre-training and Amazon Nova Forge, with the recommendation to start simple and escalate only when necessary.
AWS Machine Learning describes using foundation models to simplify catalog-wide demand forecasting and then building a closed detect–decide–act loop on Databricks and Amazon Quick that compares demand surges to live supplier availability and places replenishment orders automatically, escalating to a human only when no supplier can cover a surge.
An AWS Machine Learning blog post argues that comparing models by dollars per million tokens misses production costs tied to outcomes, and shares an open-source benchmarking harness that measures cost per correct answer, agent trajectory cost, and rubric-graded deliverable quality across OpenAI models on Amazon Bedrock.
AWS Machine Learning published guidance on building and deploying MCP Apps with interactive HTML widgets using Amazon Bedrock AgentCore. Because MCP Apps is a host-agnostic standard, the same server can deliver the same rich experience across AI hosts that support the extension, such as ChatGPT and Claude.
TwelveLabs' Marengo Embed 3.0 is now generally available as an embedding model in Amazon Bedrock Knowledge Bases, enabling fully managed natural-language semantic search over video, image and audio content. AWS Machine Learning published a walkthrough showing how to build a Bedrock knowledge base powered by Marengo 3.0 and run semantic queries against media.
Amazon today announced the desktop version of its AI assistant Amazon Quick is generally available for macOS and Windows. The company also introduced a new activity feed on iOS and Android that consolidates email, calendar and CRM items; Amazon emphasizes that data stays in customers' environments and conversations remain private.
AWS Machine Learning published a how-to showing how to build an end-to-end RFI questionnaire workflow using Amazon Quick Automate: read a multi-tab RFI workbook from Amazon S3, use natural-language prompts to extract and structure the data, iterate the workflow conversationally, and write CSV output back to S3. The guide claims this approach can reduce development time from days to hours.
According to AWS Machine Learning, they developed a configurable, model-agnostic detector that uses prompts to turn any LLM on Amazon Bedrock into a PII detector; because entity types live in the prompt rather than code, the detector can adapt to new entities without retraining and (AWS reports) outperformed an off-the-shelf tool across five public corpora and nine LLM-based detectors.
An AWS Machine Learning post introduces the Agent Evaluation Metric (AEM), a decomposable, turn-level metric for evaluating multi-turn conversational agents. The post demonstrates AEM's first dimension—correctness—showing how it can pinpoint the specific turn that caused a failure and distinguish that root cause from later turns that inherited the error.
AWS Machine Learning provides a walkthrough for deploying Qwen3.8-2.4T-A95B, a 2.4‑trillion‑parameter open‑weight model, on Amazon SageMaker HyperPod using vLLM. The guide covers cluster provisioning, NVFP4 quantization, and creating an OpenAI‑compatible endpoint with built‑in reasoning, tool calling, and native MTP speculative decoding.
Allow Google Analytics to measure visits? It stays off until you choose. You can change your choice at any time.