[{"data":1,"prerenderedAt":153},["ShallowReactive",2],{"work\u002Ftubr::{\"resolve_relations\":\"case_study_group.case_studies,work_history.roles,role.case_studies\",\"version\":\"published\"}":3,"datasource:sectors":136,"datasource:capabilities":144},{"data":4,"headers":113},{"story":5,"cv":110,"rels":111,"links":112},{"name":6,"created_at":7,"published_at":8,"updated_at":9,"id":10,"uuid":11,"content":12,"slug":101,"full_slug":102,"sort_by_date":103,"position":104,"tag_list":105,"is_startpage":40,"parent_id":106,"meta_data":103,"group_id":107,"first_published_at":8,"release_id":103,"lang":108,"path":103,"alternates":109,"default_full_slug":103,"translated_slugs":103},"Tubr: from chat to feed","2026-11-05T22:46:41.269Z","2026-11-05T23:19:35.917Z","2026-11-05T23:19:36.418Z",238506808432275,"01c51406-a99d-42f9-8168-55e65c8d95d4",{"_uid":13,"body":14,"title":6,"period":86,"my_role":87,"sectors":88,"summary":90,"question":91,"component":92,"work_type":93,"capabilities":94,"organisation":100},"66112eb7-9166-4541-aefd-e076869736f5",[15,28,33,55,64,77],{"_uid":16,"text":17,"heading":26,"component":27},"06d4b7cd-06e1-467f-a8ce-a5c9e9f4a9e9",{"type":18,"content":19},"doc",[20],{"type":21,"content":22},"paragraph",[23],{"text":24,"type":25},"When I joined, the app was chat-first: you asked it questions about your forecast. In my first six weeks I rebuilt and replatformed the customer experience, in time for a launch in the UK and US.","text","The starting point","text_section",{"_uid":29,"text":30,"label":31,"component":32},"8f5d3561-ef7d-4808-bc1e-56106ee78f97","I spent a lot of time talking to people who work in hospitality about their day and how a prediction would change what they did. People running a venue are busy. They want to glance at what's coming, not start a conversation to find out.","What I heard","insight",{"_uid":34,"heading":35,"options":36,"decision":48,"component":54},"2df1ad9e-ae33-40ac-9c89-57d66b5c52e8","Chat or feed?",[37,43],{"_uid":38,"name":39,"chosen":40,"component":41,"description":42},"ba1cbf97-01ef-4ed4-8627-c0841810b3d0","Chat-first",false,"option_item","The existing approach. Flexible, but the user has to know what to ask.",{"_uid":44,"name":45,"chosen":46,"component":41,"description":47},"f84af6c1-d9cc-44fe-9945-3ebc8ec3bc5f","Information-first feed",true,"Push insights to the user as a feed, with a dashboard view of what's coming.",{"type":18,"content":49},[50],{"type":21,"content":51},[52],{"text":53,"type":25},"I chose the feed. It puts the information in front of people instead of making them ask for it. Later I built it into a system of record: people can comment on forecasts and look back at what actually happened.","options_considered",{"_uid":56,"text":57,"heading":63,"component":27},"96dd32d6-f7ae-4035-ac52-d4c793c91ae6",{"type":18,"content":58},[59],{"type":21,"content":60},[61],{"text":62,"type":25},"I designed the data model for a multi-tenant feed across companies and their outlets, so the right people see the right insights by user, role, outlet and company. Per-user state lives apart from the generated content, so regenerating an insight can't wipe out someone's comments.","Making the feed work",{"_uid":65,"text":66,"heading":76,"component":27},"7fda646b-1224-4683-9a91-8976a6a82c05",{"type":18,"content":67},[68,72],{"type":21,"content":69},[70],{"text":71,"type":25},"When I arrived, deploying the infrastructure meant running Terraform from someone's laptop. I've been moving deployment into the merge process. The newer parts of the infrastructure now build a preview environment for each pull request, and tear down any that haven't merged after seven days.",{"type":21,"content":73},[74],{"text":75,"type":25},"I lead a lead data scientist and an ML engineer, and still ship front-end and back-end code myself. I've moved the team onto AI-native ways of working, including agentic workflows for parallel development and bug fixing, and added PostHog instrumentation so we can see how the app is used.","How the team ships",{"_uid":78,"text":79,"heading":85,"component":27},"1f2acc3b-e804-47cf-ae42-d550bb863744",{"type":18,"content":80},[81],{"type":21,"content":82},[83],{"text":84,"type":25},"Still early. We're working towards launch, partners are coming on board, and the first customer found us on their own.","Where it is","2026–present","Fractional Chief Product Officer",[89],"hospitality","Tubr uses machine learning to predict trade for hospitality businesses. I moved the app from chat-first to a feed of insights, then changed how the team ships.","Should a forecasting app be something you talk to, or something you glance at?","case_study","zero_to_one",[95,96,97,98,99],"product-direction","product-design","front-end","technical-leadership","ai-assisted-delivery","Tubr","tubr","work\u002Ftubr",null,-70,[],238506020350277,"a893f6b3-21df-4494-9faf-4e06a435497c","default",[],1793922325,[],[],{"cache-control":114,"connection":115,"content-encoding":116,"content-type":117,"date":118,"etag":119,"referrer-policy":120,"sb-be-version":121,"server":122,"strict-transport-security":123,"transfer-encoding":124,"vary":125,"via":126,"x-amz-cf-id":127,"x-amz-cf-pop":128,"x-cache":129,"x-content-type-options":130,"x-frame-options":131,"x-permitted-cross-domain-policies":132,"x-request-id":133,"x-runtime":134,"x-xss-protection":135},"max-age=0, public, s-maxage=604800, stale-if-error=3600","keep-alive","gzip","application\u002Fjson; charset=utf-8","Thu, 05 Nov 2026 23:46:01 GMT","W\u002F\"76bf196e02173da0b3920615ecb99a59\"","strict-origin-when-cross-origin","5.1002.0","nginx","max-age=31536000","chunked","Origin,Accept-Encoding","1.1 0ef7f058b6800e1f8ea404e35f354a4e.cloudfront.net (CloudFront)","_SFPZNgwxGUG0la1x8OIwWfKHcZnSkZuag3xRwx-4YNHLDgK9mRHJw==","LAX54-P1","Miss from cloudfront","nosniff","SAMEORIGIN","none","662b58cb-2b01-4937-b749-9d193ede90b2","0.017259","0",{"manufacturing":137,"ict-managed-services":138,"hospitality":139,"print-ecommerce":140,"healthcare":141,"architecture-property":142,"home-services":143},"Manufacturing","ICT managed services","Hospitality","Print & e-commerce","Healthcare","Architecture & property","Home services",{"service-design":145,"product-design":146,"front-end":147,"product-direction":148,"facilitation":149,"technical-leadership":150,"ai-assisted-delivery":151,"discovery":152},"Service design","Product design","Front-end development","Product direction","Facilitation","Technical leadership","AI-assisted delivery","Discovery",1793922360906]