{"schema_version":"newruntime-week-v0.1","type":"week_in_review","generated_at":"2026-11-18T18:00:02.626Z","key":"2026-w39","label":"Sep 21–27, 2026","start":"2026-09-21","end":"2026-09-27","count":108,"routes":{"html":"https://newruntime.com/weeks/2026-w39/","json":"https://newruntime.com/weeks/2026-w39.json","markdown":"https://newruntime.com/weeks/2026-w39.md","index":"https://newruntime.com/weeks.json","previous":"https://newruntime.com/weeks/2026-w38.json","next":"https://newruntime.com/weeks/2026-w40.json"},"note_routes":{"html":"https://newruntime.com/posts/{slug}/","markdown":"https://newruntime.com/posts/{slug}.md","json":"https://newruntime.com/posts/{slug}.json"},"summary":{"label":"New Runtime synthesis","headline":"AI agents are moving from capable interfaces to governed, cost-aware operating systems","paragraphs":["The clearest shift is from agents that answer or generate to systems that execute inside real workflows. Claude’s browser, CRM and small-business tools, Microsoft’s Copilot redesign, and production patterns for QA, triage and document handling all make permissions, routing and approvals part of the product. In parallel, model and runtime changes make longer work more economical through context reuse, lower-cost options and inference optimization. The advantage moves from access to a strong model toward teams that can safely connect it to useful work.","That shift does not make general models universal substitutes. Notes on OCR, specialized parsing, diarization and local classification show why narrow components can remain cheaper or more dependable for bounded tasks; agentic systems still need evidence, evaluation and human review. Meanwhile, browser and repository access raise the cost of mistakes: persistent agents can cross boundaries, and containment depends on the surrounding system, not just the prompt. More plug-ins, skills and marketplaces widen choice but also make integration and governance more crowded.","Earlier weeks emphasized control planes, operating models and managed agent runs; this week continues that thread and makes it more concrete at the workflow and distribution layers. For builders and buyers, compare cost per completed task and the strength of permission, evidence and rollback controls—not token price or model claims alone—and test specialized components where correctness matters."],"key_stories":[{"slug":"claude-opus-5-5-gpt-6-sol-gpt-b807a1","title":"Claude Opus 5.5, GPT-6 Sol, GPT-6 Luna, and a new price war","why":"The release cluster frames competition around the economics of agentic work, not just model rankings."},{"slug":"claude-for-small-business-launches-new-workflows-integrations-39f4e2","title":"Claude for Small Business launches new workflows, integrations, and training programs","why":"Scheduled workflows with connected systems and approval gates show agents becoming operational layers."},{"slug":"claude-in-chrome-is-generally-available-analysis-b82a25","title":"Claude in Chrome is generally available","why":"A paid browser agent with conditional autonomy makes execution a mainstream product surface while increasing the importance of access controls."},{"slug":"exposing-a-github-token-in-a-public-repository-de8732","title":"Exposing a GitHub token in a public repository","why":"A persistent coding agent’s token exposure shows how ordinary repository access can become a cross-boundary risk."},{"slug":"document-ocr-is-not-getting-commoditized-analysis-f1b2ed","title":"Document OCR is Not Getting Commoditized","why":"Specialized document engines can retain accuracy and cost advantages over general frontier models."},{"slug":"what-a-task-costs-on-opus-5-5-0b85af","title":"What a task costs on Opus 5.5","why":"Measuring cost by completed agentic work shifts buyer comparisons beyond token prices."},{"slug":"daily-ai-brief-sept-26-microsoft-biggest-copilot-a0fd1e","title":"Microsoft is rolling out a Copilot redesign centered on Home, Code and Autopilot","why":"The Copilot redesign reflects the broader move from assistant interface toward governed execution."}],"source_note_count":108,"generated_at":"2026-10-26T12:53:11.627Z"},"topics":[{"slug":"claude","label":"Claude","count":11,"notes":[{"slug":"cad-skill-claude-code-skill-for-parametric-3d-5a4a12","title":"cad-skill: Claude Code skill for parametric 3D-printable model generation with CadQuery","day":"2026-09-26"},{"slug":"build-plugins-for-claude-with-the-directory-submission-fe2989","title":"Build plugins for Claude with the directory submission portal","day":"2026-09-25"},{"slug":"claude-tag-now-supports-personal-connectors-in-channels-510bba","title":"Claude Tag now supports personal connectors in channels","day":"2026-09-24"},{"slug":"claude-marketplace-plugins-and-connectors-products-and-agents-9096d6","title":"Claude Marketplace: plugins and connectors, products and agents, and service partners","day":"2026-09-23"},{"slug":"claude-marketplace-plugins-and-connectors-products-and-agents-90e32f","title":"Claude Marketplace: plugins and connectors, products and agents, and service partners","day":"2026-09-23"},{"slug":"claude-opus-5-5-gpt-6-sol-gpt-b807a1","title":"Claude Opus 5.5, GPT-6 Sol, GPT-6 Luna, and a new price war","day":"2026-09-23"},{"slug":"salesforce-in-claude-analysis-09ab86","title":"Anthropic released Salesforce in Claude in beta","day":"2026-09-22"},{"slug":"claude-for-financial-advisors-analysis-313ee1","title":"Claude for Financial Advisors","day":"2026-09-22"},{"slug":"claude-for-small-business-launches-new-workflows-integrations-39f4e2","title":"Claude for Small Business launches new workflows, integrations, and training programs","day":"2026-09-22"},{"slug":"claude-in-chrome-is-generally-available-analysis-b82a25","title":"Claude in Chrome is generally available","day":"2026-09-22"},{"slug":"projects-redesigned-from-folder-to-conversation-analysis-884164","title":"Projects redesigned: from folder to conversation","day":"2026-09-22"}]},{"slug":"agents","label":"Agents","count":7,"notes":[{"slug":"mvueh-an-enigma-message-recovered-analysis-c35fea","title":"MVUEH — An Enigma message recovered","day":"2026-09-27"},{"slug":"univer-workspace-an-open-source-office-workspace-where-fdfe2d","title":"univer-workspace: An open-source Office workspace where people and AI agents create, collaborate, and review…","day":"2026-09-26"},{"slug":"agentic-conversational-video-intelligence-built-on-aws-analysis-75928c","title":"Agentic conversational video intelligence built on AWS","day":"2026-09-24"},{"slug":"introducing-exa-agent-ultra-lists-without-limits-analysis-ad81c5","title":"Introducing Exa Agent Ultra – Lists without Limits","day":"2026-09-24"},{"slug":"how-an-ai-agent-automates-qa-for-the-f8e9ca","title":"How an AI Agent Automates QA for the Cerebras Cloud Console","day":"2026-09-23"},{"slug":"nvidia-nemotron-3-diarization-real-time-speaker-labels-d73a05","title":"NVIDIA Nemotron 3 Diarization: real-time speaker labels at a cent per audio hour","day":"2026-09-23"},{"slug":"liteparse-updates-september-2026-analysis-cc0eab","title":"LiteParse Updates: September 2026","day":"2026-09-22"}]},{"slug":"models","label":"Models","count":7,"notes":[{"slug":"self-play-pretraining-with-zero-data-https-arxiv-0618c0","title":"Self-Play Pretraining with Zero Data","day":"2026-09-26"},{"slug":"document-ocr-is-not-getting-commoditized-analysis-f1b2ed","title":"Document OCR is Not Getting Commoditized","day":"2026-09-25"},{"slug":"flux-3-action-a-7b-world-action-model-f8ea8f","title":"FLUX 3 Action: A 7B World Action Model for Robot Control","day":"2026-09-25"},{"slug":"search-capability-leaderboard-24-models-ranked-on-web-8a7b71","title":"Search Capability Leaderboard: 24 models ranked on web search","day":"2026-09-25"},{"slug":"your-transformer-can-hold-two-thoughts-at-once-c49646","title":"Your Transformer Can Hold Two Thoughts at Once: Evidence of Linear Superposition in LLMs","day":"2026-09-25"},{"slug":"how-to-train-your-own-jev-for-17-9a96f0","title":"How to train your own Jev for $17","day":"2026-09-24"},{"slug":"canary-rollouts-upgrade-models-in-production-without-downtime-1f9940","title":"Canary rollouts: upgrade models in production without downtime","day":"2026-09-23"}]},{"slug":"inference","label":"Inference","count":6,"notes":[{"slug":"hitting-a-billion-tokens-per-minute-on-one-f36638","title":"Hitting a billion tokens per minute on one GPU by combining a query planner and an inference engine","day":"2026-09-27"},{"slug":"review-of-russia-s-serial-modul-neural-chip-134e51","title":"Review of russia s serial modul neural chip and deployments","day":"2026-09-27"},{"slug":"nvidia-model-optimizer-analysis-6652fd","title":"NVIDIA’s Model Optimizer repository now includes an end-to-end tutorial for Qwen3.6-35B-A3B combining NVFP4","day":"2026-09-24"},{"slug":"leejet-stable-diffusion-cpp-analysis-f6ea19","title":"stable-diffusion.cpp has added Day-0 support for Qwen-Image-2.1, according to the project’s repository","day":"2026-09-24"},{"slug":"how-to-serve-trillions-of-tokens-for-trillion-6b3dbd","title":"How to serve trillions of tokens for trillion","day":"2026-09-23"},{"slug":"taming-the-ml-firehose-scaling-feature-consistency-analysis-1b9ea4","title":"Taming the ML Firehose: Scaling Feature Consistency","day":"2026-09-23"}]},{"slug":"coding-agents","label":"Coding agents","count":5,"notes":[{"slug":"kong-the-world-s-first-agentic-reverse-engineer-177054","title":"kong: The world's first agentic reverse engineer","day":"2026-09-27"},{"slug":"exposing-a-github-token-in-a-public-repository-de8732","title":"Exposing a GitHub token in a public repository","day":"2026-09-26"},{"slug":"airi-open-nethack-planning-community-analysis-7a306f","title":"Airi open nethack planning community","day":"2026-09-25"},{"slug":"skills-a-collection-of-docker-skills-for-ai-0e0bfd","title":"skills: A collection of Docker skills for AI coding agents to help them build, test, debug, and optimize…","day":"2026-09-25"},{"slug":"use-open-weight-models-as-your-ai-coding-3209eb","title":"Use open weight models as your AI coding agent with Amazon Bedrock","day":"2026-09-24"}]},{"slug":"gemini","label":"Gemini","count":4,"notes":[{"slug":"google-launches-gemini-3-8-live-with-live-b93daf","title":"Google launches Gemini 3.8 Live with Live Avatar Google launched Gemini 3.8 Live Avatar for enterprise, adding near real-time video personas to conversational AI.","day":"2026-09-26"},{"slug":"gemini-3-8-tts-playground-analysis-e5135b","title":"Gemini 3.8 TTS Playground","day":"2026-09-24"},{"slug":"introducing-gemini-3-8-live-with-live-avatar-9ca5da","title":"Introducing Gemini 3.8 Live with Live Avatar","day":"2026-09-24"},{"slug":"gemini-3-8-flash-tts-and-gemini-3-4270e0","title":"Gemini 3.8 Flash TTS and Gemini 3.8 Flash","day":"2026-09-23"}]},{"slug":"generative-media","label":"Generative media","count":4,"notes":[{"slug":"ai-generated-interactive-pixel-art-animation-and-browser-a013d9","title":"Ai generated interactive pixel art animation and browser automation demo","day":"2026-09-27"},{"slug":"deploying-real-time-personalized-speech-with-qwen3-tts-598bc4","title":"Deploying real-time personalized speech with Qwen3-TTS on Amazon SageMaker AI","day":"2026-09-26"},{"slug":"viggle-qwen-image-2-1-viggle-turbo-analysis-815bca","title":"Viggle released Qwen-Image-2.1-viggle-turbo v0.2.1, a runtime LoRA adapter for Qwen/Qwen-Image-2.1","day":"2026-09-26"},{"slug":"a-kernel-centric-path-to-real-time-video-83985c","title":"A kernel-centric path to real-time video generation on Trainium","day":"2026-09-25"}]},{"slug":"github","label":"GitHub","count":4,"notes":[{"slug":"push-images-from-github-actions-to-vercel-container-4c4328","title":"Push images from GitHub Actions to Vercel Container Registry without stored credentials.","day":"2026-09-25"},{"slug":"ai-powered-fuzzing-with-the-github-security-lab-5bbfca","title":"AI-powered fuzzing with the GitHub Security Lab Taskflow Agent","day":"2026-09-24"},{"slug":"developers-want-more-efficient-software-here-s-what-cffb4d","title":"Developers want more efficient software. Here’s what over 1000 GitHub users told us they need.","day":"2026-09-23"},{"slug":"rendering-huge-pull-requests-in-the-github-copilot-3da14c","title":"Rendering huge pull requests in the GitHub Copilot app","day":"2026-09-23"}]},{"slug":"agent-workflows","label":"Agent workflows","count":3,"notes":[{"slug":"compass-is-coming-to-the-cloud-analysis-bea975","title":"Compass is coming to the cloud","day":"2026-09-25"},{"slug":"dspy-3-4-0-out-with-1-native-4439f5","title":"DSPy 3.4.0 out, with: (1) native support for Jev and System One models - in the timeless DSPy syntax.","day":"2026-09-25"},{"slug":"bring-trusted-data-into-parallel-with-data-connectors-bca538","title":"Bring trusted data into Parallel with Data Connectors","day":"2026-09-23"}]},{"slug":"ai-infrastructure","label":"AI infrastructure","count":3,"notes":[{"slug":"scaling-moe-reinforcement-learning-on-amazon-eks-with-6e49ed","title":"Scaling MoE reinforcement learning on Amazon EKS with EFA and DeepEP with 40% more throughput","day":"2026-09-26"},{"slug":"avito-recommendation-architecture-analysis-5ad956","title":"AvitoTech reports that it redesigned the collaborative recommendation path for Avito’s home page because a","day":"2026-09-25"},{"slug":"every-byte-counts-arcv3-and-the-case-for-b00788","title":"Every byte counts: ARCv3 and the case for cross","day":"2026-09-24"}]},{"slug":"anthropic","label":"Anthropic","count":3,"notes":[{"slug":"introducing-claude-opus-5-5-analysis-47f772","title":"Introducing Claude Opus 5.5","day":"2026-09-25"},{"slug":"coding-sessions-are-longer-and-use-more-context-a12d56","title":"Coding sessions are longer and use more context. Claude Opus 5.5 is built with that in mind.","day":"2026-09-24"},{"slug":"what-a-task-costs-on-opus-5-5-0b85af","title":"What a task costs on Opus 5.5","day":"2026-09-22"}]},{"slug":"mcp","label":"MCP","count":3,"notes":[{"slug":"build-a-multi-account-ai-agent-with-agentcore-b891d7","title":"Build a multi-account AI agent with AgentCore Gateway and MCP","day":"2026-09-25"},{"slug":"parse-extract-classify-now-each-with-its-own-dc0f04","title":"Parse, Extract, Classify — now each with its own MCP","day":"2026-09-25"},{"slug":"parse-extract-classify-now-each-with-its-own-9c14cb","title":"Parse, Extract, Classify — now each with its own MCP","day":"2026-09-25"}]},{"slug":"observability","label":"Observability","count":3,"notes":[{"slug":"lan-orangutan-lan-orangutan-is-a-lightweight-network-4fcacd","title":"LAN-Orangutan: LAN Orangutan is a lightweight network scanner with persistent device labeling, multi-network…","day":"2026-09-26"},{"slug":"datasette-1-0a41-analysis-0dd7a4","title":"Datasette 1.0a41 adds OpenTelemetry support, contributed by Alec Garcia, according to the release announcement","day":"2026-09-25"},{"slug":"jev-as-a-judge-in-langfuse-evaluators-analysis-09f0c3","title":"Jev-as-a-judge in Langfuse Evaluators","day":"2026-09-23"}]},{"slug":"retrieval","label":"Retrieval","count":3,"notes":[{"slug":"merge-picks-parallel-search-for-universal-context-layer-b225e3","title":"Merge picks Parallel Search for Universal Context Layer","day":"2026-09-25"},{"slug":"speakr-speakr-is-a-personal-self-hosted-web-ab80ec","title":"speakr: Speakr is a personal, self-hosted web application designed for transcribing audio recordings","day":"2026-09-25"},{"slug":"pinecone-byoc-trusted-ai-knowledge-in-the-customer-a792e1","title":"Pinecone BYOC: Trusted AI Knowledge in the Customer Cloud","day":"2026-09-22"}]},{"slug":"security","label":"Security","count":3,"notes":[{"slug":"gpt-6-sol-now-leads-deepsecbench-s-pareto-6da7d4","title":"GPT-6 Sol now leads DeepSecBench's Pareto frontier: • Highest benchmark score • 80% lower run cost than the runner-up (GPT-6 Astra)","day":"2026-09-25"},{"slug":"misconfiguring-turnstile-by-skipping-backend-validation-leaves-sites-a78fbc","title":"Misconfiguring Turnstile by skipping backend validation leaves sites exposed to bots.","day":"2026-09-25"},{"slug":"the-vercel-bug-bounty-program-is-now-publicly-801278","title":"The Vercel Bug Bounty Program is now publicly available","day":"2026-09-24"}]},{"slug":"agent-harness","label":"Agent harness","count":2,"notes":[{"slug":"openrig-multi-agent-harness-for-coordinating-claude-code-72f6c8","title":"Openrig multi agent harness for coordinating claude code and codex","day":"2026-09-27"},{"slug":"dsh-tui-plugin-for-deepseek-harness-analysis-13ec3d","title":"Dsh tui plugin for deepseek harness","day":"2026-09-26"}]},{"slug":"developer-tools","label":"Developer tools","count":2,"notes":[{"slug":"commit-rewriter-0-2-analysis-201a04","title":"commit-rewriter 0.2 adds support for operating on Git branches other than the repository’s default branch","day":"2026-09-25"},{"slug":"llm-typesafe-0-1a0-analysis-24b1a3","title":"Simon Willison released llm-typesafe 0.1a0, an LLM CLI plugin that adds support for TypeSafe AI’s Jev model","day":"2026-09-22"}]},{"slug":"enterprise-ai","label":"Enterprise AI","count":2,"notes":[{"slug":"when-ai-moves-faster-than-the-organization-analysis-61d4a8","title":"When AI Moves Faster Than the Organization","day":"2026-09-27"},{"slug":"deploying-ai-from-pilot-to-production-a-practical-8a59a4","title":"Deploying AI from pilot to production: a practical blueprint for CIOs and technical leaders","day":"2026-09-22"}]},{"slug":"google","label":"Google","count":2,"notes":[{"slug":"google-beam-expands-to-new-countries-analysis-b611d2","title":"Google Beam expands to new countries","day":"2026-09-24"},{"slug":"learn-about-google-s-project-suncatcher-to-put-e594f2","title":"Learn about Google’s Project Suncatcher to put ML infrastructure in space","day":"2026-09-24"}]},{"slug":"interfaces","label":"Interfaces","count":2,"notes":[{"slug":"introducing-meta-vr-glasses-a-cinema-courtside-seat-d2b042","title":"Introducing Meta VR Glasses: A Cinema, Courtside Seat, and Workspace in Just 100 Grams","day":"2026-09-24"},{"slug":"introducing-ray-ban-meta-audio-and-more-ai-c6913c","title":"Introducing Ray-Ban Meta Audio and More AI Glasses Styles","day":"2026-09-24"}]},{"slug":"local-models","label":"Local models","count":2,"notes":[{"slug":"bonsai-2-27b-near-lossless-compression-in-a-613c4a","title":"Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint","day":"2026-09-25"},{"slug":"fastino-gliner2-5-decide-analysis-322203","title":"Fastino released GLiNER2.5-Decide, a 340M English model for operational text classification","day":"2026-09-25"}]},{"slug":"multimodal","label":"Multimodal","count":2,"notes":[{"slug":"accelerate-multimodal-rl-training-with-skyrl-on-amazon-0eb012","title":"Accelerate multimodal RL training with SkyRL on Amazon SageMaker HyperPod","day":"2026-09-26"},{"slug":"speaker-labeled-transcription-with-whisperx-on-sagemaker-ai-93ebee","title":"Speaker-labeled transcription with WhisperX on SageMaker AI","day":"2026-09-25"}]},{"slug":"sandbox","label":"Sandbox","count":2,"notes":[{"slug":"podroid-a-rootless-android-app-that-boots-alpine-7d0bcb","title":"Podroid: A rootless Android app that boots Alpine Linux: run containers (Podman/Docker/LXC) and GUI desktop…","day":"2026-09-25"},{"slug":"vercel-sandbox-now-has-memory-observability-analysis-b8617c","title":"Vercel Sandbox now has memory observability.","day":"2026-09-25"}]},{"slug":"verification","label":"Verification","count":2,"notes":[{"slug":"narrateai-production-ready-llm-quality-assurance-on-amazon-827dca","title":"NarrateAI: production-ready LLM quality assurance on Amazon Bedrock","day":"2026-09-26"},{"slug":"llm-ocr-why-the-errors-got-harder-to-4d7c51","title":"LLM OCR: Why the Errors Got Harder to Spot","day":"2026-09-25"}]},{"slug":"agent-economics","label":"Agent economics","count":1,"notes":[{"slug":"improved-token-efficiency-for-longer-agent-runs-analysis-505868","title":"Improved token efficiency for longer agent runs","day":"2026-09-25"}]},{"slug":"evals","label":"Agent evals","count":1,"notes":[{"slug":"partnering-with-accenture-on-embedded-evaluation-analysis-8e7774","title":"Partnering with Accenture on embedded evaluation","day":"2026-09-22"}]},{"slug":"agent-interfaces","label":"Agent interfaces","count":1,"notes":[{"slug":"new-in-cline-desktop-ssh-support-analysis-419006","title":"New in Cline Desktop: SSH support!","day":"2026-09-25"}]},{"slug":"agent-memory","label":"Agent memory","count":1,"notes":[{"slug":"agent-memory-with-engram-a-practical-guide-analysis-3d1cb8","title":"Agent Memory with Engram: A Practical Guide","day":"2026-09-23"}]},{"slug":"agent-ops","label":"Agent ops","count":1,"notes":[{"slug":"1-langsmith-engine-v2-2-managed-deep-agents-de58e2","title":"LangChain announced a coordinated LangSmith release spanning Engine v2, Managed Deep Agents v0.8","day":"2026-09-25"}]},{"slug":"agent-runtime","label":"Agent runtime","count":1,"notes":[{"slug":"introducing-grok-4-7-analysis-462aed","title":"SpaceXAI announced Grok 4.7 on September 21, 2026","day":"2026-09-26"}]},{"slug":"agent-security","label":"Agent security","count":1,"notes":[{"slug":"an-agent-used-dns-to-reach-an-external-1a597a","title":"An agent used DNS to reach an external chatbot","day":"2026-09-26"}]},{"slug":"skills","label":"Agent skills","count":1,"notes":[{"slug":"state-of-agent-skills-analysis-5ce9ff","title":"State of agent skills","day":"2026-09-25"}]},{"slug":"ai-adoption","label":"AI adoption","count":1,"notes":[{"slug":"what-dropbox-has-learned-from-deploying-ai-at-a9a18f","title":"What Dropbox has learned from deploying AI at company scale","day":"2026-09-23"}]},{"slug":"ai-security","label":"AI security","count":1,"notes":[{"slug":"bringing-private-processing-to-meta-ai-glasses-analysis-d9ff33","title":"Bringing Private Processing to Meta AI Glasses","day":"2026-09-24"}]},{"slug":"api-design","label":"API design","count":1,"notes":[{"slug":"liteparse-grpc-grpc-server-for-liteparse-document-parsing-1645d9","title":"liteparse-grpc: gRPC Server for LiteParse Document Parsing","day":"2026-09-25"}]},{"slug":"automation","label":"Automation","count":1,"notes":[{"slug":"building-cogsend-a-self-hosted-social-media-scheduler-ab4c62","title":"Building CogSend: A Self-Hosted Social Media Scheduler","day":"2026-09-23"}]},{"slug":"compute","label":"Compute","count":1,"notes":[{"slug":"multi-region-training-with-amazon-sagemaker-hyperpod-and-103b05","title":"Multi-Region training with Amazon SageMaker HyperPod and Qumulo","day":"2026-09-25"}]},{"slug":"enterprise-agents","label":"Enterprise agents","count":1,"notes":[{"slug":"daily-ai-brief-sept-26-microsoft-biggest-copilot-a0fd1e","title":"Microsoft is rolling out a Copilot redesign centered on Home, Code and Autopilot","day":"2026-09-26"}]},{"slug":"governance","label":"Governance","count":1,"notes":[{"slug":"llms-democratize-extraction-and-scale-malice-analysis-847cc7","title":"LLMs Democratize Extraction and Scale Malice","day":"2026-09-23"}]},{"slug":"multi-agent","label":"Multi-agent","count":1,"notes":[{"slug":"introducing-agora-2-advancing-multi-agent-world-simulation-52e8d5","title":"Introducing Agora-2: Advancing Multi-Agent World Simulation","day":"2026-09-25"}]},{"slug":"open-models","label":"Open models","count":1,"notes":[{"slug":"mimo-v2-6-pro-training-notes-analysis-7aef31","title":"MiMo-V2.6 Pro Training Notes","day":"2026-09-22"}]},{"slug":"open-source","label":"Open source","count":1,"notes":[{"slug":"contrastive-language-model-system-one-release-analysis-748b26","title":"Contrastive language model system one release","day":"2026-09-25"}]},{"slug":"orchestration","label":"Orchestration","count":1,"notes":[{"slug":"why-ai-assistants-are-slow-and-how-to-9708d9","title":"Why AI Assistants Are Slow—and How to Make Them Faster","day":"2026-09-25"}]},{"slug":"org-design","label":"Org design","count":1,"notes":[{"slug":"working-with-agentic-teammates-when-a-new-organizational-1ee338","title":"Working with Agentic ‘Teammates’: When a New Organizational Actor Collides with the Human Ecosystem of Work","day":"2026-09-27"}]},{"slug":"workflows","label":"Workflows","count":1,"notes":[{"slug":"aderant-builds-intelligent-ticket-triage-with-amazon-nova-4b8faa","title":"Aderant builds intelligent ticket triage with Amazon Nova","day":"2026-09-25"}]}]}
