[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"sanity-HXqL82eGw8XBTqUioAmnKiGfdBOD3hKAQ5cHPhp7WYE":3},{"data":4,"sourceMap":-1},{"_createdAt":5,"_id":6,"_rev":7,"_type":8,"_updatedAt":9,"blog":10,"body":25,"issue":52,"links":53,"previewText":110,"publishedAt":111,"questions":112,"sign_off":169,"slug":190,"title":193},"2026-09-01T05:33:56Z","2feec10b-eaa5-4be8-8e99-2b606b42a347","37O2kZBhAG6D7zEgzBjb2g","newsletterOverflow","2026-09-03T21:27:32Z",[11,17,21],{"_type":12,"description":13,"link":14,"promotion":15,"title":16},"newsletterOverflowStory","Ryan chats with Leo de Moura, Senior Principal Applied Scientist at AWS and the creator of the Lean language, about proving correctness in AI agents with the Lean language, how automated reasoning complements probabilistic AI models, and the use of AI for continuous code optimization.","https:\u002F\u002Fstackoverflow.blog\u002F2026\u002F08\u002F28\u002Fwhen-you-keep-ai-lean-you-keep-ai-correct\u002F?utm_source=iterable&utm_medium=email&utm_campaign=the-overflow-newsletter",null,"When you keep AI Lean, you keep AI correct ",{"_type":12,"description":18,"link":19,"promotion":15,"title":20},"Ryan is joined by Praveen Bodigutla, Principal AI Researcher at LinkedIn, to chat about the four-layer memory system his team built to give LinkedIn's hiring assistant a persistent, personalized state. ","https:\u002F\u002Fstackoverflow.blog\u002F2026\u002F08\u002F25\u002Finside-linkedin-s-cognitive-memory-agent\u002F?utm_source=iterable&utm_medium=email&utm_campaign=the-overflow-newsletter","Inside LinkedIn's cognitive memory agent for agentic personalization",{"_type":12,"description":22,"link":23,"promotion":15,"title":24},"Organizations cannot solve shadow AI with a document employees read once. They need to make responsible use easier than improvised use.","https:\u002F\u002Fstackoverflow.blog\u002F2026\u002F08\u002F24\u002Fresponsible-ai-adoption-needs-developer-workflow-design\u002F?utm_source=iterable&utm_medium=email&utm_campaign=the-overflow-newsletter","Responsible AI adoption needs developer workflow design",[26,37,44],{"_key":27,"_type":28,"children":29,"markDefs":35,"style":36},"5cdc9419dde4","block",[30],{"_key":31,"_type":32,"marks":33,"text":34},"086f917c66a8","span",[],"Getting our AI agents to act reliably and predictably can feel like a pipe dream. But who are we if not dreamers? In this issue, we’re dreaming—not hallucinating—up ways to make our AIs do what we want, whenever we want. ",[],"normal",{"_key":38,"_type":28,"children":39,"markDefs":43,"style":36},"c085b8c5a06f",[40],{"_key":31,"_type":32,"marks":41,"text":42},[],"This week, we’re chatting with the creator of Lean, Leo de Moura, about proving correctness in AI and how automated reasoning complements probabilistic models. We sit down with LinkedIn’s Praveen Bodigutla for a conversation on their new hiring assistant and the four-layer cognitive memory system powering it. Plus, we explore why responsibly designed dev workflows are the key to seeing—and defeating—shadow AI attacks. ",[],{"_key":45,"_type":28,"children":46,"markDefs":51,"style":36},"ae43ccb9a0f5",[47],{"_key":48,"_type":32,"marks":49,"text":50},"157ef3f89d4f",[],"Folks from around the web are dreaming just like we are. And some of those dreams become reality—like when one dev this week broke Auto Mode in Opus 5. Sometimes those dreams remain impossible to achieve—like actually being able to see a black hole. But there’s no use in losing sleep over any of it because we’re making all of your knowledge day dreams come true…all you need to do is head to the links below. ",[],345,[54,68,82,96],{"_key":55,"_ref":56,"_type":12,"description":57,"link":66,"promotion":15,"title":67},"0c23d7222255a12e40a1f545cb30a1ad","d55b67ca-265c-4253-b409-db9ab7fa46d5",[58],{"_key":59,"_type":28,"children":60,"markDefs":65,"style":36},"cb3a09fb43da",[61],{"_key":62,"_type":32,"marks":63,"text":64},"1f5b9c582e0d",[],"Probably did this as a way to procrastinate the actual ADHD test.",[],"https:\u002F\u002Fnullpt.rs\u002Freverse-engineering-adhd-test","Reverse engineering my ADHD test",{"_key":69,"_ref":70,"_type":12,"description":71,"link":80,"promotion":15,"title":81},"3eb686f848f52d3cb9259f115684a7dc","83901646-00a4-4f55-9911-dbdba5a3569f",[72],{"_key":73,"_type":28,"children":74,"markDefs":79,"style":36},"bc06ea6b5625",[75],{"_key":76,"_type":32,"marks":77,"text":78},"f3d65f04567c",[],"At least we all have IP addresses, right?",[],"https:\u002F\u002Fdreamstation.systems\u002Fpersonal\u002Fntppost.html","Internet centralization and the original sin of NAT",{"_key":83,"_ref":84,"_type":12,"description":85,"link":94,"promotion":15,"title":95},"1d459fbd2b3915df70a27aed8592e51e","10f9f3e0-c085-4fc1-92ba-2f4f0464e948",[86],{"_key":87,"_type":28,"children":88,"markDefs":93,"style":36},"28ad24b85149",[89],{"_key":90,"_type":32,"marks":91,"text":92},"b65760f032a2",[],"We can all agree that most AI claims are just marketing.",[],"https:\u002F\u002Fembracethered.com\u002Fblog\u002Fposts\u002F2026\u002Fbreaking-claude-code-opus-5-and-automode\u002F","Breaking Claude Code Opus 5 Auto Mode",{"_key":97,"_ref":98,"_type":12,"description":99,"link":108,"promotion":15,"title":109},"31a0d3bfc9db7ab5afdc13a29d70b0d8","44bdf632-76ec-4a5f-bbe0-f921f2cfb368",[100],{"_key":101,"_type":28,"children":102,"markDefs":107,"style":36},"f7e24a6b832b",[103],{"_key":104,"_type":32,"marks":105,"text":106},"d3058cbd590a",[],"Many people on the internet use the language part of their brain way more than the logic part.",[],"https:\u002F\u002Fnews.mit.edu\u002F2026\u002Fseparating-logic-and-language-0708","Separating logic and language","Reliable AI, breaking Opus 5, and working for exposure","2026-09-03",[113,127,141,155],{"_key":114,"_ref":115,"_type":12,"description":116,"link":125,"promotion":15,"title":126},"04dd237c5691fa91e1119c27db309f43","68a2b939-40ba-4eb2-ae30-aece6f2d089a",[117],{"_key":118,"_type":28,"children":119,"markDefs":124,"style":36},"aa7201dc3ae9",[120],{"_key":121,"_type":32,"marks":122,"text":123},"c965ca83e60f",[],"The minor inconvenience is probably worth your testers being able to actually test.",[],"https:\u002F\u002Fsoftwareengineering.stackexchange.com\u002Fquestions\u002F461431\u002Fhow-to-handle-use-cases-which-are-only-relevant-for-testers?utm_source=iterable&utm_medium=email&utm_campaign=the-overflow-newsletter","How to handle use cases which are only relevant for testers?",{"_key":128,"_ref":129,"_type":12,"description":130,"link":139,"promotion":15,"title":140},"320786fb322c62702c3650bff021b85b","c380d14a-4e7a-48bb-8a57-4e570cc37fc3",[131],{"_key":132,"_type":28,"children":133,"markDefs":138,"style":36},"1e42ca4c3db0",[134],{"_key":135,"_type":32,"marks":136,"text":137},"e1c63a8af663",[],"\"Lanterns are paid in exposure.\"",[],"https:\u002F\u002Fscifi.stackexchange.com\u002Fquestions\u002F305722\u002Fdo-the-comics-ever-address-how-much-green-lanterns-get-paid?utm_source=iterable&utm_medium=email&utm_campaign=the-overflow-newsletter","Do the comics ever address how much Green Lanterns get paid?",{"_key":142,"_ref":143,"_type":12,"description":144,"link":153,"promotion":15,"title":154},"a097a2d1a86aad868d94812e4be27a97","3bc69248-40d1-4946-b561-b74847415b55",[145],{"_key":146,"_type":28,"children":147,"markDefs":152,"style":36},"aaf31a889f85",[148],{"_key":149,"_type":32,"marks":150,"text":151},"31562336bf62",[],"Unless you've started working the job, you don't have the job yet.",[],"https:\u002F\u002Fworkplace.stackexchange.com\u002Fquestions\u002F203627\u002Fhow-likely-is-an-offer-to-be-withdrawn-after-the-employer-says-the-offer-is-bein?utm_source=iterable&utm_medium=email&utm_campaign=the-overflow-newsletter","How likely is an offer to be withdrawn after the employer says the offer is being processed but is awaiting internal sign-off?",{"_key":156,"_ref":157,"_type":12,"description":158,"link":167,"promotion":15,"title":168},"4c5012ed56b6135d543d3ea3d8f86d22","84c32b2e-8279-4094-a100-35840b7e5328",[159],{"_key":160,"_type":28,"children":161,"markDefs":166,"style":36},"063396a068c5",[162],{"_key":163,"_type":32,"marks":164,"text":165},"cfa08e56d72b",[],"You're trying to see what is, by definition, unseeable.",[],"https:\u002F\u002Fphysics.stackexchange.com\u002Fquestions\u002F875316\u002Fhow-is-the-black-disc-of-a-black-hole-visible?utm_source=iterable&utm_medium=email&utm_campaign=the-overflow-newsletter","How is the black disc of a black hole visible?",[170],{"_key":171,"_type":28,"children":172,"markDefs":186,"style":36},"c45a32def6a7",[173,177,182],{"_key":174,"_type":32,"marks":175,"text":176},"1fd607ee919c",[],"Onboard, organize, and bring your team up to speed in a jiffy. ",{"_key":178,"_type":32,"marks":179,"text":181},"d7eab5169c0d",[180],"1a54717ebd81","Try Stack Internal",{"_key":183,"_type":32,"marks":184,"text":185},"8498b2181f0e",[],".",[187],{"_key":180,"_type":188,"href":189},"link","https:\u002F\u002Fstackoverflow.co\u002Finternal\u002F?utm_source=braze&utm_medium=email&utm_campaign=the-overflow-newsletter&utm_content=footer-banner",{"_type":191,"current":192},"slug","issue-345-the-ai-we-dream-about","Issue 345: The AI we dream about"]