[{"data":1,"prerenderedAt":421},["ShallowReactive",2],{"sanity-jsOe8wflkniWJtB9jid1NJQ3i_t1yOcHdpyiPupwM-4":3,"sanity-i5EagcTUs6m8w3loxwHJ2H0lQEHSTZBxxajzgS_zeL4":139,"sanity-CkZ1FtFevLH69hYwllaMgYOjvmoCK6676hrBwRdDZBM":272,"sanity-kZWoib95zuHDDukKCe0tBtXRwc4lKNuzFQTDg4UWlzE":399},[4,82],{"description":5,"features":17,"id":80,"name":81},[6],{"_key":7,"_type":8,"children":9,"markDefs":15,"style":16},"1ca58001b386","block",[10],{"_key":11,"_type":12,"marks":13,"text":14},"0abc92708f2d","span",[],"Knowledge starts its life in all kinds of places. Ingestion lets you capture and structure those everyday nuggets that might otherwise never get shared.",[],"normal",[18,35,50,65],{"description":19,"id":28,"image":29,"name":34},[20],{"_key":21,"_type":8,"children":22,"markDefs":27,"style":16},"03f07ee10449",[23],{"_key":24,"_type":12,"marks":25,"text":26},"41c29fad675f",[],"Turn your document graveyard into a structured, verified knowledge in minutes. Power high-volume migrations via our API or web interface. ",[],"high-signal q&a, not noise",{"_type":30,"asset":31},"image",{"_ref":32,"_type":33},"image-6a730f6b8ce8eae844a47a4ea78dce1976c658b9-1200x630-png","reference","High-signal Q&A, not noise",{"description":36,"id":45,"image":46,"name":49},[37],{"_key":38,"_type":8,"children":39,"markDefs":44,"style":16},"362783e090ba",[40],{"_key":41,"_type":12,"marks":42,"text":43},"9ff08cd393b6",[],"Our AI pipeline automatically converts content into structured, tagged, and confidence scored Q&A. Your experts validate high-quality content instead of authoring from scratch.",[],"ready to review, instantly",{"_type":30,"asset":47},{"_ref":48,"_type":33},"image-0e18946bb05fde52917a76bd10a8bfba9c8df83a-1200x630-png","Ready to review, instantly",{"description":51,"id":60,"image":61,"name":64},[52],{"_key":53,"_type":8,"children":54,"markDefs":59,"style":16},"fb554ec0b58d",[55],{"_key":56,"_type":12,"marks":57,"text":58},"55b81c6bb1eb",[],"Connect your Confluence Cloud spaces to transform static pages into verified Q&A pairs that are easier to find, trust, and maintain.",[],"activate confluence data",{"_type":30,"asset":62},{"_ref":63,"_type":33},"image-09f12d5aa241f1225eae59e01cc743cf89034627-1200x630-png","Activate Confluence data",{"description":66,"id":75,"image":76,"name":79},[67],{"_key":68,"_type":8,"children":69,"markDefs":74,"style":16},"9a869179ec70",[70],{"_key":71,"_type":12,"marks":72,"text":73},"d03541728e62",[],"Put your data to work. Our MCP server lets AI agents retrieve and update SME-approved answers.",[],"deploy your intelligence ",{"_type":30,"asset":77},{"_ref":78,"_type":33},"image-e7b08ad98a996d4e2d341899b444ec4cd6d9dd35-640x360-svg","Deploy your intelligence ","an appetite to learn","An appetite to learn",{"description":83,"features":91,"id":137,"name":138},[84],{"_key":85,"_type":8,"children":86,"markDefs":90,"style":16},"5f13fba9107d",[87],{"_key":88,"_type":12,"marks":89,"text":14},"b73081b4f9c0",[],[],[92,107,122],{"description":93,"id":102,"image":103,"name":106},[94],{"_key":95,"_type":8,"children":96,"markDefs":101,"style":16},"3bcb81278487",[97],{"_key":98,"_type":12,"marks":99,"text":100},"3f91e0761db5",[],"Our time-tested Q&A surfaces trusted answers to the top, making it easy for both humans and AI to parse.",[],"q&a",{"_type":30,"asset":104},{"_ref":105,"_type":33},"image-bd65ba7f613520e0111715b63451e33983fe1fb2-640x360-svg","Q&A",{"description":108,"id":117,"image":118,"name":121},[109],{"_key":110,"_type":8,"children":111,"markDefs":116,"style":16},"1d491997e228",[112],{"_key":113,"_type":12,"marks":114,"text":115},"530a77071f05",[],"Let the AI summarize multiple answers across your knowledge base into new insights.",[],"enhanced search",{"_type":30,"asset":119},{"_ref":120,"_type":33},"image-375298b436833be26e53cdb530a8a6461912b45e-640x360-svg","Enhanced Search",{"description":123,"id":132,"image":133,"name":136},[124],{"_key":125,"_type":8,"children":126,"markDefs":131,"style":16},"811e2600d292",[127],{"_key":128,"_type":12,"marks":129,"text":130},"d7b4302ef9fd",[],"Unlike static Wikis that go stale, your community is alerted when content needs to be reviewed and updated so it stays relevant and accurate.",[],"maintainable long term",{"_type":30,"asset":134},{"_ref":135,"_type":33},"image-f38885f4c33a3c946291be452ecaf83c00295e2a-640x360-svg","Maintainable long term","always served fresh","Always served fresh",[140,153,165,177,189,201,213,224,236,248,260],{"image":141,"name":144,"quote":145},{"_type":30,"asset":142},{"_ref":143,"_type":33},"image-ee116110c7bed72a264b5d1db1fe490bcff656f4-598x118-svg","Dropbox",{"_key":146,"name":147,"role":148,"slug":149,"text":152},"b53dac9435b1","Ju Shin","Product Manager",{"_type":150,"current":151},"slug","dropbox-access","I was getting so many requests around, ‘Hey, can our team use it? Can I use it? Can I get access?",{"image":154,"name":157,"quote":158},{"_type":30,"asset":155},{"_ref":156,"_type":33},"image-bf99da124792539fa71dba0a06a8f0afcef3743e-313x70-svg","Expensify",{"_key":159,"name":160,"role":161,"slug":162,"text":164},"235dc9557ade","David Drake","Engineering",{"_type":150,"current":163},"expensify-spread","As we started to use Stack Internal and saw how nice it was to have a repository of information, we started to see it spread to other teams. Our customer support team started using it, our people success team started using it, next thing we knew, we had [Slack] integrations all over the place.",{"image":166,"name":169,"quote":170},{"_type":30,"asset":167},{"_ref":168,"_type":33},"image-69c2a1bc420a5cd1e89c04677e99de975a15f4ea-390x141-svg","84.51°",{"_key":171,"name":172,"role":173,"slug":174,"text":176},"f7259142e673","Michael Carrico","Director of Data Science",{"_type":150,"current":175},"84-51-deep-technical-knowledge","Hearing someone with very deep technical knowledge talk about what’s going on under the hood; I think is something that we just wouldn’t have had exposure to otherwise.",{"image":178,"name":181,"quote":182},{"_type":30,"asset":179},{"_ref":180,"_type":33},"image-192adaf0eb64c1f5984e13e626cb93643e0a717f-200x200-svg","HP",{"_key":183,"name":184,"role":185,"slug":186,"text":188},"b83f7bd30f01","Evan Scheessele","HP Distinguished Technologist",{"_type":150,"current":187},"hp-evan-democratized-source","It's a very special case with Stack Overflow, because it's a democratized source of well-validated, upvoted and accepted answers. [It’s] a robust source of truth. That's a key piece of the puzzle.",{"image":190,"name":193,"quote":194},{"_type":30,"asset":191},{"_ref":192,"_type":33},"image-35af1691ae92f54b035fac8fd6a330b84ce5eec3-115x21-svg","IMC",{"_key":195,"name":196,"role":197,"slug":198,"text":200},"97e3012c3a39","Tyler McDougall","Development Team Lead",{"_type":150,"current":199},"imc-we-dont-have-questions-that-are-unanswered","We don’t have questions that are unanswered more than a couple minutes. It just doesn’t happen. The user adoption and engagement with the platform has been very impressive.",{"image":202,"name":205,"quote":206},{"_type":30,"asset":203},{"_ref":204,"_type":33},"image-8a3c9051aee79bb036cf12f1503c1820b59b6562-87x53-svg","The Flex Co.",{"_key":207,"name":208,"role":209,"slug":210,"text":212},"6a3e1400e5f6","Morgan Jones","Director of Engineering",{"_type":150,"current":211},"flex-onboarding","Stack Internal saves us time on a day to day basis and also a huge amount when onboarding a new developer.",{"image":214,"name":217,"quote":218},{"_type":30,"asset":215},{"_ref":216,"_type":33},"image-38e279ddad6af76a085ed1867285bfe9c77af6f4-144x49-svg","Elastic",{"_key":219,"name":220,"role":209,"slug":221,"text":223},"a2648fe50355","Suyog Rao",{"_type":150,"current":222},"elastic-engineers-solve","Engineers should help solve the hardest questions, the unknowns, where being familiar with how the product was built is essential. But we don’t want to keep answering solved problems over and over again. That’s where Stack Overflow really helps.",{"image":225,"name":228,"quote":229},{"_type":30,"asset":226},{"_ref":227,"_type":33},"image-7795d036597430c7825f03a40b35ab53b92c39b3-102x55-svg","Box",{"_key":230,"name":231,"role":232,"slug":233,"text":235},"7fb642ed17b6","Eddie Flaiser","Head of Engineering",{"_type":150,"current":234},"box-slack","If you have a question, if you have some clarification that you need, instead of overwhelming everyone on Slack or an email, there's a queue. Questions are created, somebody is monitoring the questions and responding, and this has improved productivity across the board.",{"image":237,"name":240,"quote":241},{"_type":30,"asset":238},{"_ref":239,"_type":33},"image-e34b76bff7838f2c5bf6d1ff82615fe2649c18e4-85x18-svg","Intuit",{"_key":242,"name":243,"role":244,"slug":245,"text":247},"bb18190b128c","Rocio Montes","Staff Software Engineer",{"_type":150,"current":246},"intuit-sme","You see areas where you might have the answer and you can find solutions from subject matter experts you have never met. You are able to move faster and collaborate on code that you hadn't worked on before.",{"image":249,"name":252,"quote":253},{"_type":30,"asset":250},{"_ref":251,"_type":33},"image-dcf95975a6471089f24fd082c239ed22d3671f65-477x97-svg","Bloomberg",{"_key":254,"name":255,"role":256,"slug":257,"text":259},"2af09a9f334c","El Wieschaus","Head of Product Development Experience",{"_type":150,"current":258},"bloomberg-how-we-learn","Stack Overflow is a really critical part of how we learn, how we understand systems at Bloomberg. I literally used it this morning.",{"image":261,"name":264,"quote":265},{"_type":30,"asset":262},{"_ref":263,"_type":33},"image-a96001cdcd0d4a56b88c6452e77b0a6efe387a9d-192x41-svg","Microsoft",{"_key":266,"name":267,"role":268,"slug":269,"text":271},"7f4a85c96f82","Laura MacLeod","Program Manager, Developer Services Division",{"_type":150,"current":270},"microsoft-integrations","With the integrations between Microsoft Teams and Stack Internal, I’m excited to have one place that I can see my question within an area that I already do my work today.",[273,285,325,337,349,368,387],{"_key":274,"answer":275,"question":284},"23cec18bc522",[276],{"_key":277,"_type":8,"children":278,"markDefs":283,"style":16},"baf5d9be231b",[279],{"_key":280,"_type":12,"marks":281,"text":282},"e90e36d0a953",[],"Stack Internal’s Ingestion feature pre-populates the platform with verified technical knowledge before users are onboarded. By converting existing scattered PDFs and Markdown files into structured Q&A, it ensures the first users to search the platform find immediate, relevant answers – driving usage and adoption from Day 1.",[],"How does Ingestion solve the ”cold start“ problem?",{"_key":286,"answer":287,"question":324},"75faf0cab2b5",[288,305],{"_key":289,"_type":8,"children":290,"markDefs":304,"style":16},"c873bf10d054",[291,295,300],{"_key":292,"_type":12,"marks":293,"text":294},"9cdb042f7262",[],"Ingestion is a subscription-based add-on. All Stack Internal Enterprise sites receive a base allotment of ",{"_key":296,"_type":12,"marks":297,"text":299},"e49ad2d2c7d0",[298],"strong","100 Knowledge Objects",{"_key":301,"_type":12,"marks":302,"text":303},"b99108975433",[]," per month (measured by human approved Q&A pairs) so users can experience the power of the Ingestion tool.",[],{"_key":306,"_type":8,"children":307,"markDefs":320,"style":16},"9c434e413c05",[308,311,316],{"_key":301,"_type":12,"marks":309,"text":310},[],"For higher volume needs, organizations can purchase premium tiered add-ons. For more information, see the ",{"_key":312,"_type":12,"marks":313,"text":315},"afbe10194b39",[314],"b568c85da51b","Ingestion Subscription and Billing",{"_key":317,"_type":12,"marks":318,"text":319},"a62e2a94832d",[]," article.",[321],{"_key":314,"_type":322,"href":323},"link","https:\u002F\u002Fsupport.stackenterprise.co\u002Fsupport\u002Fsolutions\u002Farticles\u002F22000295194-ingestion-subscription-and-billing","How is Ingestion priced?",{"_key":326,"answer":327,"question":336},"75ba9189622f",[328],{"_key":329,"_type":8,"children":330,"markDefs":335,"style":16},"a0e5912f0ed0",[331],{"_key":332,"_type":12,"marks":333,"text":334},"17a80befd217",[],"Customers can import content through three primary channels: manual file uploads within the web interface, programmatic ingestion via a REST API endpoint, or the Confluence Cloud connector, which establishes a direct link between Confluence spaces and Stack Internal to automatically ingest and convert pages into structured Q&A.",[],"What options are available for importing content to Stack Internal?",{"_key":338,"answer":339,"question":348},"a51755364e94",[340],{"_key":341,"_type":8,"children":342,"markDefs":347,"style":16},"fb8acc21957f",[343],{"_key":344,"_type":12,"marks":345,"text":346},"8a46da5b2a73",[],"Customers can ingest Confluence pages, PDFs, plain text like Markdown & HTML files, images (.jpeg, .jpg, .png, .bmp, .heif, .tiff) and Microsoft 365 files (.docx, .xlsx, .pptx).",[],"What file types are supported?",{"_key":350,"answer":351,"question":367},"17ab8e0c6ba2",[352,360],{"_key":353,"_type":8,"children":354,"markDefs":359,"style":16},"03ae7c162751",[355],{"_key":356,"_type":12,"marks":357,"text":358},"e981f801a6d5",[],"Upon importing, a duplication check runs to determine if an identical or similar document already exists in the Stack Internal instance. If a duplicate is detected, it’s skipped to ensure only net-new knowledge is sent to the review queue where human experts validate the content.",[],{"_key":361,"_type":8,"children":362,"markDefs":366,"style":16},"43d6bac6f2d5",[363],{"_key":356,"_type":12,"marks":364,"text":365},[],"The Confluence Cloud connector also ensures only new or updated pages are ingested to prevent overlap.",[],"How does Ingestion handle duplicate content?",{"_key":369,"answer":370,"question":386},"b0197011bfad",[371,379],{"_key":372,"_type":8,"children":373,"markDefs":378,"style":16},"19beb069bd99",[374],{"_key":375,"_type":12,"marks":376,"text":377},"92c6d6f08cab",[],"Ingestion’s AI pipeline intelligently chunks, cleans, and converts raw text into structured, atomic Q&A pairs. These posts are automatically tagged, mapped to users, and scored for confidence before being routed to experts for final review. ",[],{"_key":380,"_type":8,"children":381,"markDefs":385,"style":16},"616348be8564",[382],{"_key":375,"_type":12,"marks":383,"text":384},[],"By converting fragmented, unstructured content into verified units of knowledge, institutional data is optimized for both human discovery and AI retrieval.",[],"How does the Ingestion AI pipeline work?",{"_key":388,"answer":389,"question":398},"79a7d38119e0",[390],{"_key":391,"_type":8,"children":392,"markDefs":397,"style":16},"a13f234c2509",[393],{"_key":394,"_type":12,"marks":395,"text":396},"370c0fad587d",[],"No. Customer data is not used to train large language models (LLMs), and external LLM providers do not have access to your data. All ingestion and AI-enhancement occurs within a secure, logically isolated environment In Azure to keep your data completely private.",[],"Does Stack Overflow use customer data to train LLMs?",[400,407,414],{"_id":401,"preface":402,"publishedAt":403,"slug":404,"title":406},"b63e62d7-d78d-4957-98ca-f5c00e08ed3c","The hidden costs of building an internal AI context layer—and why your agents are failing without it.","2026-04-29T00:54:21.693Z",{"_type":150,"current":405},"the-hidden-costs-of-building-an-internal-ai-context-layer","The hidden costs of building an internal AI context layer",{"_id":408,"preface":409,"publishedAt":410,"slug":411,"title":413},"e74572d5-a167-4f88-8ab8-88b5df48ad51","Devs don’t trust AI tools that don’t work. Community-validated data could change that.","2026-03-09T20:00:56.091Z",{"_type":150,"current":412},"why-community-validated-data-matters-more-than-ever","Why community-validated data matters more than ever",{"_id":415,"preface":416,"publishedAt":417,"slug":418,"title":420},"c60fbe6f-ffb3-4720-84de-c867f9c60167","Our bidirectional MCP server connects trusted Stack Internal knowledge to the IDE, LLM, or agent platform of your choice. This guide provides examples of prompts that we hope will help make your next big project and team handoff a breeze.","2025-12-18T10:10:00.000-05:00",{"_type":150,"current":419},"search-and-curate-trusted-knowledge-with-the-stack-internal-mcp-server","Guide: Search and curate trusted knowledge in flow with the Stack Internal MCP server",1777465362174]