[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"article-google-ax-resumable-agent-runtime-en":3,"article-related-google-ax-resumable-agent-runtime-en":32,"series-industry-36d8e7b0-ce1b-4870-b1ee-fae9ae5e2356":75},{"id":4,"slug":5,"title":6,"content":7,"summary":8,"source":9,"source_url":10,"author":11,"image_url":12,"cover_image":12,"category":13,"language":14,"translated_content":11,"related_article_id":15,"keywords":16,"key_takeaways":24,"views":28,"created_at":29,"published_at":30,"topic_cluster_id":31},"36d8e7b0-ce1b-4870-b1ee-fae9ae5e2356","google-ax-resumable-agent-runtime-en","Google AX turns agent runs into resumable jobs","\u003Cp data-speakable=\"summary\">\u003Ca href=\"\u002Ftag\u002Fgoogle\">Google\u003C\u002Fa> AX is a distributed \u003Ca href=\"\u002Ftag\u002Fagent\">agent\u003C\u002Fa> runtime for resumable, auditable agent runs.\u003C\u002Fp>\n\u003Cp>Google's \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fax\">AX\u003C\u002Fa> repo shows a runtime built for agent jobs that can pause, recover, and keep state across distributed components. The project says it is in early development and already has 1.7k stars.\u003C\u002Fp>\n\u003Ch2>1. Resumable execution\u003C\u002Fh2>\n\u003Cp>AX is built around the idea that an agent run should survive interruptions. If a client disconnects, the runtime can replay missed events from the last sequence number instead of restarting the whole conversation.\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781950658549-zc18.png\" alt=\"Google AX turns agent runs into resumable jobs\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\n\u003Cp>That matters when agent work takes time or depends on several steps. The CLI examples in the repo show both fresh runs and resumed runs, including cases where a conversation continues with new input after a failure.\u003C\u002Fp>\n\u003Cul>\n  \u003Cli>\u003Ccode>ax exec --conversation ... --resume\u003C\u002Fcode>\u003C\u002Fli>\n  \u003Cli>\u003Ccode>ax exec --last-seq 12 --resume\u003C\u002Fcode>\u003C\u002Fli>\n  \u003Cli>Resume by conversation ID\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2>2. Event logging and auditability\u003C\u002Fh2>\n\u003Cp>AX keeps execution state in an event log, with a single controller managing writes. That design gives the runtime a clear record of what happened and makes it easier to inspect or replay agent activity later.\u003C\u002Fp>\n\u003Cp>The repo frames this as a control and audit feature, not just a storage detail. User calls, agent calls, tools, and \u003Ca href=\"\u002Ftag\u002Fskills\">skills\u003C\u002Fa> all flow through the controller, which gives teams one place to trace execution.\u003C\u002Fp>\n\u003Cul>\n  \u003Cli>Single-writer architecture\u003C\u002Fli>\n  \u003Cli>Durable event log\u003C\u002Fli>\n  \u003Cli>Controller-based coordination\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2>3. Isolated agents, tools, and skills\u003C\u002Fh2>\n\u003Cp>AX treats agents, tools, and skills as isolated actors that can run separately from one another. The overview diagram shows a controller routing work to remote agents, tools, and an environment with built-in skills.\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781950655822-jpfg.png\" alt=\"Google AX turns agent runs into resumable jobs\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\n\u003Cp>That split is useful \u003Ca href=\"\u002Fnews\u002Fai-coding-assistant-roi-measured-en\">when you\u003C\u002Fa> want to keep workloads apart for safety, reliability, or scaling. It also fits systems where different tools are exposed through RPC or \u003Ca href=\"\u002Ftag\u002Fmcp\">MCP\u003C\u002Fa>-style services instead of being bundled into one process.\u003C\u002Fp>\n\u003Cul>\n  \u003Cli>Remote agents over gRPC\u003C\u002Fli>\n  \u003Cli>Tools via MCP servers\u003C\u002Fli>\n  \u003Cli>Environment and skills as separate actors\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2>4. Custom agents and CLI control\u003C\u002Fh2>\n\u003Cp>AX ships with a command-line interface that can execute built-in agents or custom ones. The repo includes examples for running a remote agent server, registering it in \u003Ccode>ax.yaml\u003C\u002Fcode>, and then invoking it through \u003Ccode>ax exec\u003C\u002Fcode>.\u003C\u002Fp>\n\u003Cp>For developers, that means AX is not only a runtime spec. It is also a practical interface for testing agent behavior, wiring in your own agent IDs, and resuming work from the same conversation state.\u003C\u002Fp>\n\u003Ccode>ax exec --server localhost:8494 --input \"Hello agents!\"\nax exec --agent coding --input \"Write a simple HTTP server in Python\"\nax serve --config my-config.yaml\u003C\u002Fcode>\n\u003Ch2>5. Kubernetes-first deployment\u003C\u002Fh2>\n\u003Cp>The project says AX is natively supported on Agent Substrate on Kubernetes and recommends that setup for production. It is also described as compute-agnostic, so the core runtime is not tied to one deployment shape.\u003C\u002Fp>\n\u003Cp>That makes AX more interesting for teams planning long-running agent workloads than for one-off demos. If you need a controller, recovery behavior, and distributed actors in a production cluster, Kubernetes is clearly the path the maintainers want you to take.\u003C\u002Fp>\n\u003Cul>\n  \u003Cli>Recommended production option: Kubernetes\u003C\u002Fli>\n  \u003Cli>Works with local and server modes\u003C\u002Fli>\n  \u003Cli>Designed for small and large deployments\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2>How to decide\u003C\u002Fh2>\n\u003Cp>Pick AX if you are building agent systems that need recovery, replay, and a clear execution record. It fits teams that expect agents to run as long-lived jobs rather than short prompt-response calls.\u003C\u002Fp>\n\u003Cp>If you only need a simple local wrapper around a model, AX is probably more runtime than you want right now. If you need distributed coordination, custom agents, and a path to Kubernetes, it is worth watching closely.\u003C\u002Fp>","5 things Google AX adds to agent systems: resumable runs, event logs, isolated actors, custom agents, and Kubernetes support.","github.com","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fax",null,"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781950658549-zc18.png","industry","en","778c1685-052f-45ce-91e0-e505f6fb2d5b",[17,18,19,20,21,22,23],"Google AX","agent runtime","distributed agents","resumable execution","event log","Kubernetes","custom agents",[25,26,27],"AX is for agent runs that need recovery and replay, not just one-shot prompts.","The controller and event log are the core of its audit and resumption model.","Custom agents, remote tools, and Kubernetes support point it toward production setups.",0,"2026-06-20T10:17:16.910807+00:00","2026-06-20T10:17:16.901+00:00","b0949719-c4ab-49f9-bb89-30226a0f4923",{"tags":33,"relatedLang":34,"relatedPosts":38},[],{"id":15,"slug":35,"title":36,"language":37},"google-ax-resumable-agent-runtime-zh","Google AX 把代理執行變成可恢復工作","zh",[39,45,51,57,63,69],{"id":40,"slug":41,"title":42,"cover_image":43,"image_url":43,"created_at":44,"category":13},"34fbfa14-37e3-4a6d-929d-a33068ac14fa","anthropic-fable-5-ban-ai-investors-en","Anthropic’s Fable 5 ban hits AI investors","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781964172715-l9xl.png","2026-06-20T14:02:27.003723+00:00",{"id":46,"slug":47,"title":48,"cover_image":49,"image_url":49,"created_at":50,"category":13},"08680bc1-fdaf-427e-a361-fdc5c5851656","army-contract-opportunities-ai-air-defense-en","Army contract opportunities for AI and air defense","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781962369594-lzb2.png","2026-06-20T13:32:21.734584+00:00",{"id":52,"slug":53,"title":54,"cover_image":55,"image_url":55,"created_at":56,"category":13},"ddc0be01-5a5c-4f45-b503-46f1adeae1b7","salesforce-fin-buy-proves-deployment-speed-wins-en","Salesforce’s Fin buy proves enterprise AI now rewards deployment speed","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781961461686-mk8m.png","2026-06-20T13:17:19.51597+00:00",{"id":58,"slug":59,"title":60,"cover_image":61,"image_url":61,"created_at":62,"category":13},"4d1ab008-c937-400b-a835-2dbc5154d67d","ai-bottleneck-power-not-chips-en","AI’s real bottleneck is power, not chips","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781960559783-52up.png","2026-06-20T13:02:17.088824+00:00",{"id":64,"slug":65,"title":66,"cover_image":67,"image_url":67,"created_at":68,"category":13},"92b6d981-2649-4c4f-a3bb-539c206861f7","solana-1000-target-fantasy-not-forecast-en","Solana’s $1,000 target is fantasy, not a forecast","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781956970314-ayh5.png","2026-06-20T12:02:19.637104+00:00",{"id":70,"slug":71,"title":72,"cover_image":73,"image_url":73,"created_at":74,"category":13},"b94d42b2-d124-4f93-98f7-305f60799562","wso2-600m-sale-open-source-enterprise-software-en","WSO2’s $600M sale caps a 20-year open-source run","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781951573378-b4r4.png","2026-06-20T10:32:31.01869+00:00",[76,81,86,91,96,101,106,111,116,121],{"id":77,"slug":78,"title":79,"created_at":80},"d35a1bd9-e709-412e-a2df-392df1dc572a","ai-impact-2026-developments-market-en","AI's Impact in 2026: Key Developments and Market Shifts","2026-03-25T16:20:33.205823+00:00",{"id":82,"slug":83,"title":84,"created_at":85},"5ed27921-5fd6-492e-8c59-78393bf37710","trumps-ai-legislative-framework-en","Trump's AI Legislative Framework: What's Inside?","2026-03-25T16:22:20.005325+00:00",{"id":87,"slug":88,"title":89,"created_at":90},"e454a642-f03c-4794-b185-5f651aebbaca","nvidia-gtc-2026-key-highlights-innovations-en","NVIDIA GTC 2026: Key Highlights and Innovations","2026-03-25T16:22:47.882615+00:00",{"id":92,"slug":93,"title":94,"created_at":95},"0ebb5b16-774a-4922-945d-5f2ce1df5a6d","claude-usage-diversifies-learning-curves-en","Claude Usage Diversifies, Learning Curves Emerge","2026-03-25T16:25:50.770376+00:00",{"id":97,"slug":98,"title":99,"created_at":100},"69934e86-2fc5-4280-8223-7b917a48ace8","openclaw-ai-commoditization-concerns-en","OpenClaw's Rise Raises Concerns of AI Model Commoditization","2026-03-25T16:26:30.582047+00:00",{"id":102,"slug":103,"title":104,"created_at":105},"b4b2575b-2ac8-46b2-b90e-ab1d7c060797","google-gemini-ai-rollout-2026-en","Google's Gemini AI Rollout Extended to 2026","2026-03-25T16:28:14.808842+00:00",{"id":107,"slug":108,"title":109,"created_at":110},"6e18bc65-42ae-4ad0-b564-67d7f66b979e","meta-llama4-fabricated-results-scandal-en","Meta's Llama 4 Scandal: Fabricated AI Test Results Unveiled","2026-03-25T16:29:15.482836+00:00",{"id":112,"slug":113,"title":114,"created_at":115},"bf888e9d-08be-4f47-996c-7b24b5ab3500","accenture-mistral-ai-deployment-en","Accenture and Mistral AI Team Up for AI Deployment","2026-03-25T16:31:01.894655+00:00",{"id":117,"slug":118,"title":119,"created_at":120},"5382b536-fad2-49c6-ac85-9eb2bae49f35","mistral-ai-high-stakes-2026-en","Mistral AI: Facing High Stakes in 2026","2026-03-25T16:31:39.941974+00:00",{"id":122,"slug":123,"title":124,"created_at":125},"9da3d2d6-b669-4971-ba1d-17fdb3548ed5","cursors-meteoric-rise-pressures-en","Cursor's Meteoric Rise Faces Industry Pressures","2026-03-25T16:32:21.899217+00:00"]