{"schema_version":"newruntime-week-v0.1","type":"week_in_review","generated_at":"2026-11-18T18:00:02.626Z","key":"2026-w36","label":"Aug 31 – Sep 6, 2026","start":"2026-08-31","end":"2026-09-06","count":45,"routes":{"html":"https://newruntime.com/weeks/2026-w36/","json":"https://newruntime.com/weeks/2026-w36.json","markdown":"https://newruntime.com/weeks/2026-w36.md","index":"https://newruntime.com/weeks.json","previous":"https://newruntime.com/weeks/2026-w35.json","next":"https://newruntime.com/weeks/2026-w37.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":"The bottleneck is shifting from making agents capable to making their work verifiable, controlled and portable","paragraphs":["Several notes point to the same constraint: code and agent actions are easier to generate than to trust. Anthropic’s lifecycle playbook, GitLab’s governance argument, and new evaluation work on voice, role drift and benchmark leakage all suggest that review, context and evidence now determine whether more output can ship. The shift reaches beyond coding: scoped connectors and policy-bound execution make admissibility part of capability, not an afterthought.","At the same time, execution is becoming a deployment choice rather than a fixed cloud default. Cursor’s customer-managed machines, isolated sandboxes, ChatGPT Work’s local-versus-cloud boundary and NVIDIA’s local inference routing give buyers more control over where work runs, while making portability and operational continuity more consequential. Cheap search and custom inference can lower unit costs, but they do not remove the burden of verifying results or managing credentials and data.","Persistent memory and shared workspaces promise less repeated setup, yet durable context also raises the stakes of accuracy, permissions and deletion; Anthropic’s distinction between deleting stored data and stopping replies makes that tension concrete. For builders and buyers, prioritize end-to-end evaluation, scoped access and a tested fallback across execution environments before expanding agent autonomy; watch whether those controls keep pace with the new volume of generated work."],"key_stories":[{"slug":"why-code-verification-matters-more-than-ever-in-the-age-of-ai","title":"Why code verification matters more than ever in the age of AI","why":"The generation-review mismatch identifies verification as the growing cost center rather than code production."},{"slug":"anthropic-ai-native-sdlc-playbook","title":"Anthropic AI-Native SDLC Playbook","why":"The lifecycle framing shows that agent gains require reorganizing review and ownership, not just adding code generation."},{"slug":"role-anchor-exposes-role-drift-in-compound-llm-pipelines","title":"Role Anchor exposes role drift in compound LLM pipelines","why":"Role drift shows why end-task accuracy alone can conceal unsafe or unreliable behavior inside agent pipelines."},{"slug":"vercel-connect-gives-apps-and-agents-scoped-connectors","title":"Vercel Connect gives apps and agents scoped connectors","why":"Task-scoped, short-lived tokens turn access control into a runtime property instead of relying on stored credentials."},{"slug":"cursor-self-hosted-machines-runtime-boundary","title":"Cursor Moves Agent Execution Into Customer-Managed Machines","why":"Moving execution onto customer-managed machines makes infrastructure control and provider portability operational concerns."},{"slug":"chatgpt-work-makes-execution-location-a-product-boundary","title":"ChatGPT Work Makes Execution Location a Product Boundary","why":"Making local versus cloud execution an explicit workflow boundary shows that placement is becoming a product decision."},{"slug":"claude-data-lifecycle-and-deletion","title":"Claude data lifecycle and deletion","why":"Separating data deletion from stopping replies exposes a practical gap between memory controls and agent behavior."}],"source_note_count":45,"generated_at":"2026-10-26T12:53:11.627Z"},"topics":[{"slug":"agents","label":"Agents","count":20,"notes":[{"slug":"claude-commerce-agents-operating-blueprint","title":"Claude Commerce Agents Package Shopping and Merchant Workflows","day":"2026-09-05"},{"slug":"a-rosetta-stone-for-ai-benchmarks","title":"A Rosetta Stone for AI benchmarks","day":"2026-09-02"},{"slug":"analyze-stripe-data-with-ai","title":"Analyze Stripe data with AI","day":"2026-09-02"},{"slug":"anti-slop-repo-for-agent-output-quality-control","title":"Anti-slop repo for agent-output quality control","day":"2026-09-02"},{"slug":"claude-remote-control","title":"Claude Remote Control","day":"2026-09-02"},{"slug":"cursor-explains-git-at-any-scale","title":"Cursor explains Git at any scale","day":"2026-09-02"},{"slug":"figmatrace-dataset-for-figma-design-workflows","title":"FigmaTrace dataset for Figma design workflows","day":"2026-09-02"},{"slug":"is-agentic-website-readiness-scoring","title":"Is Agentic website readiness scoring","day":"2026-09-02"},{"slug":"liquid-ai-agent-loops","title":"Liquid AI agent loops","day":"2026-09-02"},{"slug":"claude-academy-and-anthropic-teaching-approach","title":"Claude Academy and Anthropic teaching approach","day":"2026-09-01"},{"slug":"measuring-benchmark-optimization-in-speech-recognition","title":"Measuring benchmark optimization in speech recognition","day":"2026-09-01"},{"slug":"why-code-verification-matters-more-than-ever-in-the-age-of-ai","title":"Why code verification matters more than ever in the age of AI","day":"2026-09-01"},{"slug":"anthropic-ai-native-sdlc-playbook","title":"Anthropic AI-Native SDLC Playbook","day":"2026-08-31"},{"slug":"claude-accelerates-protein-design","title":"Claude accelerates protein design","day":"2026-08-31"},{"slug":"deepgram-flux-for-real-time-voice-agents","title":"Deepgram Flux for real-time voice agents","day":"2026-08-31"},{"slug":"gitlab-on-what-changes-when-code-is-abundant","title":"GitLab on what changes when code is abundant","day":"2026-08-31"},{"slug":"jalapeno-first-inference-results","title":"Jalapeno first inference results","day":"2026-08-31"},{"slug":"langfuse-v4-cli-and-evaluator-templates","title":"Langfuse v4, CLI and evaluator templates","day":"2026-08-31"},{"slug":"mistral-agentic-search","title":"Mistral Agentic Search","day":"2026-08-31"},{"slug":"restrict-where-claude-tag-operates","title":"Restrict where Claude Tag operates","day":"2026-08-31"}]},{"slug":"agent-security","label":"Agent security","count":11,"notes":[{"slug":"agent-lightning-v1-0","title":"Agent Lightning v1.0","day":"2026-09-01"},{"slug":"cloudflare-spectre-attack-against-workers","title":"Cloudflare Spectre attack against Workers","day":"2026-09-01"},{"slug":"full-bandwidth-transformers","title":"Full-bandwidth Transformers","day":"2026-09-01"},{"slug":"how-to-evaluate-live-and-voice-agents-in-adk","title":"How to evaluate live and voice agents in ADK","day":"2026-09-01"},{"slug":"policy-algebra-for-trust-preserving-agent-execution","title":"Policy Algebra for trust-preserving agent execution","day":"2026-09-01"},{"slug":"role-anchor-exposes-role-drift-in-compound-llm-pipelines","title":"Role Anchor exposes role drift in compound LLM pipelines","day":"2026-09-01"},{"slug":"vercel-connect-gives-apps-and-agents-scoped-connectors","title":"Vercel Connect gives apps and agents scoped connectors","day":"2026-09-01"},{"slug":"build-zero-trust-ai-agents-with-google-s-adk","title":"Build zero-trust AI agents with Google's ADK","day":"2026-08-31"},{"slug":"claude-data-lifecycle-and-deletion","title":"Claude data lifecycle and deletion","day":"2026-08-31"},{"slug":"cloudflare-bot-preference-sync","title":"Cloudflare Bot Preference Sync","day":"2026-08-31"},{"slug":"parallel-search-fast","title":"Parallel Search Fast prices web search for cheap models","day":"2026-08-31"}]},{"slug":"coding-agents","label":"Coding agents","count":4,"notes":[{"slug":"cursor-self-hosted-machines-runtime-boundary","title":"Cursor Moves Agent Execution Into Customer-Managed Machines","day":"2026-09-05"},{"slug":"vercel-sandbox-cursor-cloud-agent-execution","title":"Vercel Sandbox Turns Cursor Cloud Agents Into Isolated Worker Requests","day":"2026-09-05"},{"slug":"local-coding-agents-are-becoming-shared-workspaces","title":"Local Coding Agents Are Becoming Shared Workspaces","day":"2026-08-31"},{"slug":"model-access-portability-is-an-operational-resilience-layer","title":"Model Access Portability Is an Operational Resilience Layer","day":"2026-08-31"}]},{"slug":"agent-memory","label":"Agent memory","count":2,"notes":[{"slug":"mem0-agent-memory-layer","title":"mem0 agent memory layer","day":"2026-09-02"},{"slug":"warp-agent-memory","title":"Warp Agent Memory","day":"2026-09-01"}]},{"slug":"generative-media","label":"Generative media","count":2,"notes":[{"slug":"lyria-35-gemini-music-product-surface","title":"Lyria 3.5 Moves Music Generation Into Gemini's Everyday Product Surface","day":"2026-09-05"},{"slug":"creative-ai-is-splitting-generation-into-production-primitives","title":"Creative AI Is Splitting Generation Into Production Primitives","day":"2026-08-31"}]},{"slug":"models","label":"Models","count":2,"notes":[{"slug":"claude-fable-mythos-safeguard-split","title":"Claude Fable and Mythos Make the Model Picker a Policy Surface","day":"2026-09-05"},{"slug":"gemini-agentic-video-understanding-workflow","title":"Google Puts Agentic Video Understanding Into the Gemini Model Surface","day":"2026-09-05"}]},{"slug":"ai-adoption","label":"AI adoption","count":1,"notes":[{"slug":"mckinsey-ai-management-operating-loop","title":"AI Management Moves From Tool Adoption To Operating Capability","day":"2026-09-06"}]},{"slug":"chatgpt-work","label":"chatgpt-work","count":1,"notes":[{"slug":"chatgpt-work-makes-execution-location-a-product-boundary","title":"ChatGPT Work Makes Execution Location a Product Boundary","day":"2026-08-31"}]},{"slug":"inference","label":"Inference","count":1,"notes":[{"slug":"nvidia-pair-local-inference-router","title":"NVIDIA PAIR Makes Local Inference a Routing Problem","day":"2026-09-05"}]},{"slug":"mcp","label":"MCP","count":1,"notes":[{"slug":"cohere-ate-agentic-task-ecosystem","title":"Cohere ATE Measures What Developers Have Turned Into Agent Tools","day":"2026-09-05"}]}]}
