[{"data":1,"prerenderedAt":1396},["ShallowReactive",2],{"footer-vacatures-en":3,"hub-ai-laten-bouwen-en":4,"content-by-keys-offers-en-build-integrate,ai-prototype-sprint,vibe-code-kickstart":78,"content-by-keys-useCases-en-ai-agent-laten-bouwen,interne-ai-assistent-laten-ontwikkelen,bedrijfsproces-automatiseren-met-ai,ai-toevoegen-aan-bestaande-software,eigen-data-en-kennis-gebruiken-met-ai,documenten-en-mails-automatisch-verwerken,ai-prototype-laten-bouwen,product-goed-opzetten-voor-ai-development,agentic-workflows-bestaande-software,tools-vervangen-door-maatwerkflow":478,"content-by-keys-buyerTriggers-en-starten-met-ai,vastgelopen-met-ai":1242,"content-by-keys-cases-en-":1375,"why-9to5-en":1376},[],{"id":5,"title":6,"approach":7,"body":22,"cta":29,"description":26,"extension":32,"hero":33,"key":35,"label":36,"meta":37,"navigation":38,"path":39,"proof":40,"relations":52,"seo":69,"stem":72,"summary":73,"visuals":74,"__hash__":77},"buyerTriggers_en\u002Fbuyer-triggers\u002Fai-laten-bouwen.md","We know what we want. Now it needs to be built.",{"title":8,"steps":9},"From a clear scope to something that really works.",[10,13,16,19],{"title":11,"text":12},"Define the scope and success criteria","We make it clear what needs to work, who it is for and when it is good enough to go live.\n",{"title":14,"text":15},"Set up the architecture and integrations","We determine how AI, data, existing software, permissions and integrations should work together technically.\n",{"title":17,"text":18},"Build and test","We implement the difficult part and test more than the happy path, including reliability, errors and human oversight where needed.\n",{"title":20,"text":21},"Bring it into production and hand over","We launch it properly, document the important decisions and make sure your team can carry on independently.\n",{"type":23,"value":24,"toc":25},"minimark",[],{"title":26,"searchDepth":27,"depth":27,"links":28},"",2,[],{"title":30,"body":31},"Know what needs to be built?","Explain what you need and which systems it should work with. We'll quickly tell you how to build it properly and what a logical first step would be.\n","md",{"intro":34},"You know what needs to be built. We help you build it properly, connect it to your existing systems and get it running reliably in production, using Claude, ChatGPT\u002FOpenAI, Codex or MCP where appropriate. We step in, build, document and hand over, without an open-ended staffing commitment.\n","ai-laten-bouwen","Build with AI",{},true,"\u002Fbuyer-triggers\u002Fai-laten-bouwen",{"intro":41,"points":42},"AI is only really useful when it works with existing software, data and business processes. We build that as software engineering, not as a standalone demo.\n",[43,46,49],{"title":44,"text":45},"Integrations are central to our work","We treat APIs, databases, permissions and existing systems as core parts of the solution, not extras around the AI.\n",{"title":47,"text":48},"Reliable beyond the happy path","We also test errors, edge cases and human oversight where needed.\n",{"title":50,"text":51},"Production and handover are part of the job","The work is done when the solution is live, documented and manageable for your team.\n",{"useCaseKeys":53,"offerKeys":64,"caseKeys":68},[54,55,56,57,58,59,60,61,62,63],"ai-agent-laten-bouwen","interne-ai-assistent-laten-ontwikkelen","bedrijfsproces-automatiseren-met-ai","ai-toevoegen-aan-bestaande-software","eigen-data-en-kennis-gebruiken-met-ai","documenten-en-mails-automatisch-verwerken","ai-prototype-laten-bouwen","product-goed-opzetten-voor-ai-development","agentic-workflows-bestaande-software","tools-vervangen-door-maatwerkflow",[65,66,67],"build-integrate","ai-prototype-sprint","vibe-code-kickstart",[],{"title":70,"description":71},"AI agent development and integration | 9to5 software","Have senior engineers build an AI agent, MCP integration, assistant or automation with Claude, ChatGPT\u002FOpenAI or Codex. From scope to production.\n","buyer-triggers\u002Fai-laten-bouwen","The problem and direction are clear. Now it needs to be built properly, integrated and brought into production.\n",{"hero":75,"proof":76,"approach":76},"sq-foundation-build","ls-system-assemble","ShT4oNEyuY5KbwbXvOuC2vEuCmku084FYvyfv89UAyM",[79,219,349],{"id":80,"title":81,"body":82,"catalogue":86,"cta":97,"description":26,"extension":32,"fit":102,"hero":106,"key":65,"meta":109,"navigation":38,"next":110,"outcome":122,"path":134,"price":135,"process":138,"proof":153,"relations":167,"scope":176,"seo":206,"stem":209,"summary":210,"summaryPanel":211,"visuals":216,"__hash__":218},"offers_en\u002Foffers\u002Fbuild-integrate.md","Build & Integrate",{"type":23,"value":83,"toc":84},[],{"title":26,"searchDepth":27,"depth":27,"links":85},[],{"audience":87,"promise":88,"deliverables":89,"note":95,"visual":96},"For organisations that know what needs to be built and want 9to5 to build and integrate it professionally.\n","Not an AI demo, but something that works with your real software, data, processes and people.\n",[90,91,92,93,94],"Architecture and production-ready implementation","Integrations with existing software and data","Permissions, error handling and human oversight","Tests, monitoring and deployment where relevant","Documentation and handover to your team","We define the scope together first, then agree a fixed price for the result.\n","ls-system-connect",{"title":98,"body":99,"primaryCta":100,"local":101},"Do you know what needs to be built?","Share the desired outcome and your existing technical context with us. If the scope is clear, we turn it into a fixed engagement.\n","Discuss what you need built","AI engineering and software implementation from Delft, for organisations in Delft, The Hague, Rotterdam and the surrounding area.\n",{"title":103,"intro":104,"useCaseKeys":105},"When is this offer a fit?","When the desired outcome is clear enough to scope a professional implementation.\n",[54,55,56,57],{"headline":107,"intro":108,"primaryCta":100},"Not an AI demo, but something that works with your real software, data and people.","The problem and the desired direction are clear. Now it has to be built properly. We take on architecture, AI, integrations and production-grade engineering as one whole and hand it back over afterwards.\n",{},{"title":111,"items":112},"What happens next?",[113,116,119],{"title":114,"text":115},"You take it over","That is what we work towards. We don't want you to keep needing us.\n",{"title":117,"text":118},"A new bounded question comes up later","Then we scope it again instead of automatically keeping capacity running.\n",{"title":120,"text":121,"offerKey":66},"Still too much technical uncertainty?","Then an AI Prototype Sprint or Reality Check before Build & Integrate may be wiser.\n",{"title":123,"items":124},"Afterwards you have...",[125,128,131],{"title":126,"text":127},"Working software instead of a demo","The agreed functionality works end to end within the real technical environment.\n",{"title":129,"text":130},"Reliable integration with what is already there","Data, systems, permissions and human processes are part of the solution.\n",{"title":132,"text":133},"A product you can hand over","Architecture, code and key choices are documented well enough to continue independently.\n","\u002Foffers\u002Fbuild-integrate",{"label":136,"type":137},"Fixed price after scoping","scoped",{"title":139,"steps":140},"How it works.",[141,144,147,150],{"title":142,"text":143},"Scope + success criteria","We make exact enough what needs to work to agree on responsibility and price.\n",{"title":145,"text":146},"Architecture + integrations","We determine how AI, data, existing software, permissions and users come together.\n",{"title":148,"text":149},"Build + test","We implement the hard part and also test errors, edge cases and reliability.\n",{"title":151,"text":152},"Go live + hand over","We bring the solution to production in a controlled way, document it and hand it over.\n",{"title":154,"intro":155,"points":156,"caseKeys":166},"Why should you believe this?","AI only becomes truly valuable when it works together with existing software, data and business processes. That is simply serious software engineering.\n",[157,160,163],{"title":158,"text":159},"The same people advise and build","Advice and execution sit with the same technical people.\n",{"title":161,"text":162},"Integrations are core work","We don't treat APIs, databases, auth and existing systems as an afterthought around the AI.\n",{"title":164,"text":165},"Production and handover are part of it","The work isn't done when the demo works, but when the solution is manageable.\n",[],{"buyerTriggerKeys":168,"useCaseKeys":170,"relatedOfferKeys":171,"caseKeys":175},[35,169],"vastgelopen-met-ai",[54,55,56,57,58,59,62,63],[172,66,173,174],"ai-reality-check","vibe-code-rescue-audit","rescue-sprint",[],{"title":177,"intro":178,"activities":179,"deliverables":190,"exclusions":199},"What is included in this offer?","The scope is defined up front. Then you get a fixed price for the agreed result.\n",{"intro":180,"items":181},"The exact work differs per engagement, but we take responsibility for the complete technical result.\n",[182,183,184,185,186,187,188,189],"Making scope and success criteria concrete","Designing the architecture","Implementing AI functionality or an agent flow","Building integrations with existing software and data","Setting up authentication, authorisation and data flows properly","Tests, evals, monitoring and error handling where relevant","Deployment and production setup","Documentation and technical handover",{"intro":191,"items":192},"The deliverables are defined per scope and are part of the fixed price.\n",[193,194,195,196,197,198],"Working agreed functionality","Production-grade implementation","Required integrations","Relevant tests and quality checks","Technical documentation","Deployment and handover as scoped",{"intro":200,"items":201},"No unbounded consultancy and no feature list that keeps growing afterwards.\n",[202,203,204,205],"Functionality outside the agreed scope","Ongoing retainer as a standard component","External model, API and hosting costs","Additional work without a new explicit scope and price",{"title":207,"description":208},"Have AI built and integrated | Build & Integrate | 9to5 software","Have an AI agent, internal tool, MCP integration or AI feature built with Claude or ChatGPT\u002FOpenAI and professionally integrated with your software, data and processes.\n","offers\u002Fbuild-integrate","You know what needs to be built. We build it professionally, connect it to your existing software and make sure it can be handed over production-ready.\n",{"qualifier":212,"result":213,"duration":214,"nextStep":215},"Fixed price after joint scoping","Working, integrated software your team can take over","Set in the scope","Handover to your team, or further development together",{"hero":217,"proof":75,"process":96},"sq-connect-nodes","jlXsSL_a-hghGaOxvEHbZ2ARDUkoaAlfXaEkKLzQpSQ",{"id":220,"title":221,"body":222,"catalogue":226,"cta":236,"description":26,"extension":32,"fit":241,"hero":246,"key":66,"meta":249,"navigation":38,"next":250,"outcome":261,"path":272,"price":273,"process":276,"proof":290,"relations":303,"scope":309,"seo":336,"stem":339,"summary":340,"summaryPanel":341,"visuals":346,"__hash__":348},"offers_en\u002Foffers\u002Fai-prototype-sprint.md","AI Prototype Sprint",{"type":23,"value":223,"toc":224},[],{"title":26,"searchDepth":27,"depth":27,"links":225},[],{"audience":227,"promise":228,"deliverables":229,"note":235,"visual":76},"For a specific AI idea whose key technical assumptions need to be validated first.\n","From one clear use case to a working end-to-end prototype that lets you really test the technology.\n",[230,231,232,233,234],"Technical architecture for one use case","Working end-to-end prototype","Up to two key integrations","Validation of key assumptions and reliability","Demo, documentation and advice on moving to production","A working technical flow, not a clickable mock-up.",{"title":237,"body":238,"primaryCta":239,"local":240},"Prove it works technically first?","Share the use case with us. We'll help you choose the smallest prototype that actually proves something important.\n","Start an AI Prototype Sprint","AI prototyping by senior engineers from Delft, for organisations in the region.\n",{"title":103,"intro":242,"useCaseKeys":243},"When the idea is clear enough to build, but still needs to be proven before you go to production.\n",[60,244,245],"ai-betrouwbaar-genoeg","ai-toevoegen-aan-bestaand-product",{"headline":247,"intro":248,"primaryCta":239},"Prove that it works before you turn it into a complete product.","You roughly know what you want, but not yet whether the technology behaves in practice as you hope. In one bounded Sprint we build the key flow and test the assumptions that determine whether investing further is wise.\n",{},{"title":111,"items":251},[252,255,258],{"title":253,"text":254,"offerKey":65},"The prototype proves the direction","Then Build & Integrate can deliver the production version.\n",{"title":256,"text":257},"The outcome calls for adjustment","Then you have learned concretely which technical route doesn't work, without building a complete product first.\n",{"title":259,"text":260,"offerKey":173},"The prototype has become production itself","If an experiment already has real users and needs professionalising, the Rescue route is a better fit.\n",{"title":123,"items":262},[263,266,269],{"title":264,"text":265},"A working end-to-end prototype","The key user flow actually works with the chosen AI technology and integrations.\n",{"title":267,"text":268},"Technical assumptions are tested","You know better where quality, reliability and integrations are or aren't feasible.\n",{"title":270,"text":271},"A decision about production","You know what it takes to turn the prototype into a serious product or production component.\n","\u002Foffers\u002Fai-prototype-sprint",{"label":274,"type":275},"€6,950 excl. VAT","fixed",{"title":139,"steps":277},[278,281,284,287],{"title":279,"text":280},"Define use case + success criterion","We pin down which technical assumption this Sprint needs to prove.\n",{"title":282,"text":283},"Set up architecture + flow","We choose the shortest technical route that validates the real problem.\n",{"title":285,"text":286},"Build + evaluate","We deliver the end-to-end flow and test where the solution is fragile.\n",{"title":288,"text":289},"Demo + production advice","We show what works, what is still uncertain and what a next phase needs.\n",{"title":154,"intro":291,"points":292,"caseKeys":302},"A good prototype is more than a pretty click-through demo. It answers the technical question that decides whether the concept can work.\n",[293,296,299],{"title":294,"text":295},"We test the real thing","We test real integrations, data flows and model behaviour where relevant.\n",{"title":297,"text":298},"Small enough to learn","We keep you from building half a product before the key assumption is proven.\n",{"title":300,"text":301},"Ready for a real follow-up decision","Continuing, adjusting or stopping can all be a good outcome.\n",[],{"buyerTriggerKeys":304,"useCaseKeys":306,"relatedOfferKeys":307,"caseKeys":308},[305,35],"starten-met-ai",[60],[172,65,173],[],{"title":177,"intro":310,"activities":311,"deliverables":322,"exclusions":329},"One clear use case, one primary flow and a controlled amount of technology around it.\n",{"intro":312,"items":313},"We build enough to test the real technical risk, without building the complete end product right away.\n",[314,315,316,317,318,319,320,321],"Sharpening scope and success criteria","Defining the technical architecture","Building a working prototype","Up to two key external integrations","Delivering one primary end-to-end flow","Configuring the model, agent or instructions","Validating the key technical assumptions","Evaluating reliability at a high level",{"intro":323,"items":324},"A tangible prototype plus enough technical clarity for a follow-up decision.\n",[231,325,326,327,197,328],"Technical architecture direction","Validated core flow","Demo of the result","Advice for production and next steps",{"intro":330,"items":331},"A Prototype Sprint proves the core. Production hardening is a separate step.\n",[332,333,334,335],"Full production guarantees or SLAs","Extensive scalability and operationalisation","Unbounded integrations or multiple main flows","Full product development beyond the agreed use case",{"title":337,"description":338},"AI Prototype Sprint | Working AI prototype | 9to5 software","Turn one AI concept into a working end-to-end prototype, for example with Claude or OpenAI. Includes architecture, integrations, validation and production advice.\n","offers\u002Fai-prototype-sprint","From one concrete AI idea to a working end-to-end prototype that lets you actually test the key technical assumptions.\n",{"qualifier":342,"result":343,"duration":344,"nextStep":345},"Fixed price · one concrete AI idea","A working end-to-end prototype with tested technical assumptions","One bounded Sprint of about three weeks","A reasoned decision about production, for example via Build & Integrate",{"hero":75,"proof":347,"process":76},"sq-process-steps","WOLcKM4lFt5312Z6kkvWsCyd0jk_equXhLoXFeu1UMg",{"id":350,"title":351,"body":352,"catalogue":356,"cta":366,"description":26,"extension":32,"fit":371,"hero":375,"key":67,"meta":378,"navigation":38,"next":379,"outcome":390,"path":401,"price":402,"process":404,"proof":418,"relations":431,"scope":436,"seo":465,"stem":468,"summary":469,"summaryPanel":470,"visuals":475,"__hash__":477},"offers_en\u002Foffers\u002Fvibe-code-kickstart.md","Vibe Code Kickstart",{"type":23,"value":353,"toc":354},[],{"title":26,"searchDepth":27,"depth":27,"links":355},[],{"audience":357,"promise":358,"deliverables":359,"note":365,"visual":75},"For teams that want to keep building themselves with Claude Code, Codex, Cursor, Lovable, Bolt or similar AI tools.\n","We put the right foundation in place. Then you or your team can keep building independently.\n",[360,361,362,363,364],"Stack, architecture and project structure","Data, authentication and environments where needed","Deployment, logging and error handling","Test strategy and AI coding guardrails","One validated end-to-end flow + handover","A professional foundation. Then you build on it.",{"title":367,"body":368,"primaryCta":369,"local":370},"Building yourself, but starting right?","Tell us what you want to make and how you want to keep building afterwards.\n","Start a Vibe Code Kickstart","From Delft we help teams and entrepreneurs in the region get AI-assisted development off the ground in a technically sound way.\n",{"title":103,"intro":372,"useCaseKeys":373},"When you want to build yourself with AI tools, but don't want to gamble with the technical foundation.\n",[374,61],"vibe-coding-goed-beginnen",{"headline":376,"intro":377,"primaryCta":369},"We lay the foundation properly. Then you keep building.","AI coding tools make building software more accessible and faster. But the wrong technical foundation gets expensive later. We set up one product or codebase professionally, validate a working end-to-end flow and then hand it over to you.\n",{},{"title":111,"items":380},[381,384,387],{"title":382,"text":383},"You keep building","That's the usual route. The Kickstart lets you pick up speed on your own afterwards.\n",{"title":385,"text":386,"offerKey":174},"Get stuck on something specific later?","Then a Rescue Sprint can resolve one concrete blocker.\n",{"title":388,"text":389,"offerKey":173},"Has the project gone off the rails technically?","Then a Vibe Code Rescue Audit first restores the overview.\n",{"title":123,"items":391},[392,395,398],{"title":393,"text":394},"A professional technical foundation","Architecture, project structure, data, deployment and access are set up deliberately.\n",{"title":396,"text":397},"Guardrails for AI-assisted development","Tests, conventions and project instructions help coding agents work within the right boundaries.\n",{"title":399,"text":400},"A product you can take forward yourself","One end-to-end path works and the key technical choices are documented.\n","\u002Foffers\u002Fvibe-code-kickstart",{"label":403,"type":275},"€4,950 excl. VAT",{"title":139,"steps":405},[406,409,412,415],{"title":407,"text":408},"Sharpen product + way of working","We determine what you want to build yourself, with which tools and which technical requirements come with it.\n",{"title":410,"text":411},"Lay the foundation","We set up architecture, data, auth, deployment and project structure properly.\n",{"title":413,"text":414},"Add guardrails","We set up tests, conventions and instructions so AI tools make better technical choices.\n",{"title":416,"text":417},"Validate + hand over","We build one end-to-end path, document the foundation and make sure you can continue on your own.\n",{"title":154,"intro":419,"points":420,"caseKeys":430},"We have nothing against vibe coding. With a good engineering foundation you can build much more yourself before you need senior help again.\n",[421,424,427],{"title":422,"text":423},"Senior software engineering as the starting point","The foundation is set up as if the product will actually be used later.\n",{"title":425,"text":426},"Focused on AI-assisted development","The structure helps both people and coding agents understand the codebase.\n",{"title":428,"text":429},"Dependency is not the business model","The Kickstart has succeeded when you can carry on without us.\n",[],{"buyerTriggerKeys":432,"useCaseKeys":433,"relatedOfferKeys":434,"caseKeys":435},[305,35],[374,61],[174,173],[],{"title":177,"intro":437,"activities":438,"deliverables":450,"exclusions":458},"One bounded Kickstart for one product or codebase.\n",{"intro":439,"items":440},"We set up exactly the technical parts that keep speed from turning into chaos later.\n",[441,442,443,444,445,446,447,448,449],"Choosing the stack and architecture","Setting up the repository and project structure","Laying down the base data model","Authentication and authorisation where needed","Environments and deployment structure","Logging and error handling","Test strategy and technical conventions","Guardrails and instructions for AI coding tools","Validating one primary end-to-end flow",{"intro":451,"items":452},"A working technical foundation that is ready for you to build on.\n",[453,454,197,455,456,457],"Set-up codebase and project structure","Working base for deployment and environments","AI development instructions and conventions","Validated primary flow","Handover explaining the key choices",{"intro":459,"items":460},"We lay the foundation and don't build the complete product within this Kickstart.\n",[461,462,463,464],"Fully worked-out product functionality","Unbounded feature development","Ongoing technical guidance or retainer","Extensive migration of an existing complex codebase",{"title":466,"description":467},"Vibe Code Kickstart | Set up AI coding properly | 9to5 software","Building software yourself with Claude Code, Codex, Cursor, Lovable or Bolt? 9to5 sets up architecture, deployment, auth, data, tests and AI coding guardrails professionally.\n","offers\u002Fvibe-code-kickstart","We set up the technical foundation properly. After that, you or your team keep building independently with Claude Code, Codex, Cursor, Lovable, Bolt or other AI coding tools.\n",{"qualifier":471,"result":472,"duration":473,"nextStep":474},"Fixed price · one product or codebase","A professional technical foundation with one working end-to-end flow","About two weeks","You or your team continue independently with AI coding tools",{"hero":75,"proof":476,"process":76},"sq-guardrails-check","e9euJh732-H8mgX6Ag-tYTMatUDDPi69gH_31YmK9Ms",[479,556,633,710,786,862,938,1014,1090,1166],{"id":480,"title":481,"body":482,"description":26,"desiredOutcome":486,"extension":32,"hero":498,"key":54,"meta":500,"navigation":38,"path":501,"problem":502,"proof":514,"recognition":526,"recommendedApproach":541,"relations":543,"seo":548,"stem":551,"summary":552,"visuals":553,"__hash__":555},"useCases_en\u002Fuse-cases\u002Fai-agent-laten-bouwen.md","Have an AI agent built",{"type":23,"value":483,"toc":484},[],{"title":26,"searchDepth":27,"depth":27,"links":485},[],{"intro":487,"items":488},"You don't want a standalone AI demo, but an agent that performs a clear task and fits the software and processes that already exist.\n",[489,492,495],{"title":490,"text":491},"A working end-to-end flow","The agent can retrieve the relevant information, make decisions and take the right actions.\n",{"title":493,"text":494},"Control over what it may do","Permissions, validation and human oversight limit its autonomy where needed.\n",{"title":496,"text":497},"Manageable in production","Errors, model behaviour, costs and tool calls are visible enough to use the agent responsibly.\n",{"intro":499},"An agent that only gives clever answers won't do much for you. It becomes useful when it can retrieve information, make choices and safely take actions in your existing systems. We build that whole technical setup.\n",{},"\u002Fuse-cases\u002Fai-agent-laten-bouwen",{"intro":503,"causes":504},"An AI agent is more than a prompt with a few functions. It is a software system in which unpredictable models are allowed to take real actions.\n",[505,508,511],{"title":506,"text":507},"Autonomy introduces risk","A wrong model decision can have consequences outside the chat interface.\n",{"title":509,"text":510},"Integrations determine the real value","Without good access to existing data and systems, the agent remains mostly a demonstration.\n",{"title":512,"text":513},"Reliability is built around the model","Validation, retries, state, logging and human-in-the-loop are often just as important as the choice of model.\n",{"intro":515,"points":516},"9to5 software combines AI engineering with senior software engineering. So next to the smart part, we also build the technical environment you need to use an agent seriously.\n",[517,520,523],{"title":518,"text":519},"Agents and regular software","Tooling, APIs, data, auth and error handling belong to the same solution.\n",{"title":521,"text":522},"Human-in-the-loop where needed","Not every process gets better with maximum autonomy.\n",{"title":524,"text":525},"No black box left behind","We document the important choices so the system stays transferable.\n",{"intro":527,"signals":528},"The use case is often already clear. The technical challenge is running the complete workflow reliably.\n",[529,532,535,538],{"title":530,"text":531},"The agent has to do more than answer","It has to retrieve data, call systems or actually change something.\n",{"title":533,"text":534},"Several tools or systems are involved","The workflow runs across APIs, databases or existing business software.\n",{"title":536,"text":537},"Not every action may run automatically","For critical steps a user must be able to check or approve.\n",{"title":539,"text":540},"A simple prompt is not enough","State, context, error handling and tool use become part of the technical design.\n",{"intro":542},"With Build & Integrate we design and build the complete agent flow: architecture, tools, integrations, permissions, guardrails and production behaviour. When a new or uncertain agent concept first needs to be proven technically, an AI Prototype Sprint can be a more logical first step.\n",{"buyerTriggerKeys":544,"primaryOfferKey":65,"offerKeys":545,"caseKeys":546,"relatedUseCaseKeys":547},[35],[65,66],[],[62,56],{"title":549,"description":550},"AI agent development (Claude, ChatGPT, MCP) | 9to5 software","Need an AI agent for a serious business process? 9to5 builds agents with Claude or OpenAI, MCP tools and integrations with your existing software and data.\n","use-cases\u002Fai-agent-laten-bouwen","You know which task or workflow the agent has to perform. Now it needs to be designed well, connected and made reliable enough for real use.\n",{"hero":347,"problem":476,"proof":554},"ls-agentic-loop","Sz6OZV7bqm5zl00ThdeWnUL9u1nZHw5GFlz7cgVRcnE",{"id":557,"title":558,"body":559,"description":26,"desiredOutcome":563,"extension":32,"hero":575,"key":55,"meta":577,"navigation":38,"path":578,"problem":579,"proof":591,"recognition":603,"recommendedApproach":618,"relations":620,"seo":625,"stem":628,"summary":629,"visuals":630,"__hash__":632},"useCases_en\u002Fuse-cases\u002Finterne-ai-assistent-laten-ontwikkelen.md","Have an internal AI assistant developed",{"type":23,"value":560,"toc":561},[],{"title":26,"searchDepth":27,"depth":27,"links":562},[],{"intro":564,"items":565},"Employees should get to the right information faster without creating a new knowledge problem with unreliable answers.\n",[566,569,572],{"title":567,"text":568},"Answers based on your own sources","The assistant uses relevant internal documents, data and knowledge.\n",{"title":570,"text":571},"Sources stay verifiable","Users can see where important information comes from.\n",{"title":573,"text":574},"Access follows existing permissions","The assistant doesn't reveal information the user normally isn't allowed to see.\n",{"intro":576},"An internal assistant only becomes truly useful when it knows your context, can find sources and only shows information a user is allowed to see. We build the technology behind that experience.\n",{},"\u002Fuse-cases\u002Finterne-ai-assistent-laten-ontwikkelen",{"intro":580,"causes":581},"An internal AI assistant is mainly an information and integration challenge around a language model.\n",[582,585,588],{"title":583,"text":584},"Knowledge is scattered","Documents, databases and other systems each have their own structure.\n",{"title":586,"text":587},"A model doesn't know what is true","Without good retrieval and source selection, convincing-sounding nonsense can appear.\n",{"title":589,"text":590},"Permissions have to carry through","Central access to knowledge must not bypass existing security boundaries.\n",{"intro":592,"points":593},"The chat window is the easy part. Whether employees can rely on the assistant depends on how data, permissions, retrieval and AI are connected.\n",[594,597,600],{"title":595,"text":596},"Your own data as part of the design","We start with the sources and users, not with an arbitrary model.\n",{"title":598,"text":599},"Integrations are part of it","The assistant can connect to existing software instead of becoming a new information silo.\n",{"title":601,"text":602},"Reliability over magic","Sources, uncertainty and limits may stay visible to the user.\n",{"intro":604,"signals":605},"The need is usually clear when a lot of valuable knowledge does exist, but is hard to find or scattered.\n",[606,609,612,615],{"title":607,"text":608},"Employees keep searching for the same information","Answers are spread across documents, systems and colleagues.\n",{"title":610,"text":611},"A general chatbot doesn't know your context","Good answers require access to organisation-specific information.\n",{"title":613,"text":614},"Sources change regularly","The solution has to work with current knowledge without baking everything into a model.\n",{"title":616,"text":617},"Not everyone may see everything","Roles, teams or clients have different access rights.\n",{"intro":619},"With Build & Integrate we build the assistant around your real sources, access model and usage situations. If it's still uncertain whether retrieval and answer quality will work well enough, we can first validate that core in an AI Prototype Sprint.\n",{"buyerTriggerKeys":621,"primaryOfferKey":65,"offerKeys":622,"caseKeys":623,"relatedUseCaseKeys":624},[35],[65,66],[],[58,59],{"title":626,"description":627},"Have an internal AI assistant developed | 9to5 software","Need an internal AI assistant or your own ChatGPT built on your knowledge and systems? 9to5 builds secure, integrated AI assistants for serious internal use.\n","use-cases\u002Finterne-ai-assistent-laten-ontwikkelen","You want to give employees faster access to internal knowledge, documents and systems through an assistant that fits your organisation.\n",{"hero":631,"problem":217,"proof":96},"sq-knowledge-filter","_peqkPNQXiEZoQx11qqN8kY_wuZT1mvO0j1tCcY3dno",{"id":634,"title":635,"body":636,"description":26,"desiredOutcome":640,"extension":32,"hero":652,"key":56,"meta":654,"navigation":38,"path":655,"problem":656,"proof":668,"recognition":680,"recommendedApproach":695,"relations":697,"seo":702,"stem":705,"summary":706,"visuals":707,"__hash__":709},"useCases_en\u002Fuse-cases\u002Fbedrijfsproces-automatiseren-met-ai.md","Automate a business process with AI",{"type":23,"value":637,"toc":638},[],{"title":26,"searchDepth":27,"depth":27,"links":639},[],{"intro":641,"items":642},"You don't want as much AI as possible. You want less manual work, fewer errors and a process that stays technically manageable.\n",[643,646,649],{"title":644,"text":645},"Less repetitive work","Predictable or time-consuming steps are carried out automatically.\n",{"title":647,"text":648},"AI only where it adds something","Interpretation and unstructured information go to AI; fixed rules stay regular software.\n",{"title":650,"text":651},"People stay in control","Critical exceptions or decisions can deliberately stay with an employee.\n",{"intro":653},"Not every process gets better with a fully autonomous agent. Sometimes the best solution is regular automation with one smart AI step in between. We build the workflow that fits the process.\n",{},"\u002Fuse-cases\u002Fbedrijfsproces-automatiseren-met-ai",{"intro":657,"causes":658},"AI automation works best when you first technically split the process into deterministic and non-deterministic steps.\n",[659,662,665],{"title":660,"text":661},"Not every step needs AI","Using a language model for fixed rules makes processes more expensive and less predictable.\n",{"title":663,"text":664},"Exceptions determine reliability","The system has to know when continuing automatically is not wise.\n",{"title":666,"text":667},"Integrations are part of the process","The workflow has to be able to read and write in the systems employees already use today.\n",{"intro":669,"points":670},"For us AI is one tool within software automation. So we don't make a process artificially agentic when a simpler design is more reliable.\n",[671,674,677],{"title":672,"text":673},"Process before technology","The desired outcome determines which steps need AI.\n",{"title":675,"text":676},"Existing systems stay usable","Integrations prevent an automation from becoming yet another standalone system.\n",{"title":678,"text":679},"Human oversight is part of the design","We explicitly design where exceptions should go to an employee.\n",{"intro":681,"signals":682},"Processes with a lot of copying, checking, interpreting and passing on are a good fit.\n",[683,686,689,692],{"title":684,"text":685},"People move information between systems","A large part of the work consists of reading, retyping or classifying.\n",{"title":687,"text":688},"The same decision rules keep coming back","A workflow usually follows a fixed pattern with a limited number of exceptions.\n",{"title":690,"text":691},"Unstructured information blocks classic automation","Emails, documents or free text first have to be interpreted.\n",{"title":693,"text":694},"Existing software has to stay part of it","You don't want to replace all current systems to make one process smarter.\n",{"intro":696},"With Build & Integrate we design and build the complete workflow around the existing process. If there is technical uncertainty about a crucial AI step, we can validate it first in an AI Prototype Sprint.\n",{"buyerTriggerKeys":698,"primaryOfferKey":65,"offerKeys":699,"caseKeys":700,"relatedUseCaseKeys":701},[35],[65,66],[],[59,63],{"title":703,"description":704},"Automate a business process with AI | 9to5 software","Automating a business process with AI? 9to5 builds workflows that practically combine AI, existing software, data and human oversight.\n","use-cases\u002Fbedrijfsproces-automatiseren-met-ai","The process is clear and contains work that could be smarter or largely automatic. Now it has to become a reliable technical workflow.\n",{"hero":347,"problem":476,"proof":708},"ls-workflow-flow","-G_t_-Jsnlyir8qFMNtC3gxskycjna1O6PRo9qtkdak",{"id":711,"title":712,"body":713,"description":26,"desiredOutcome":717,"extension":32,"hero":729,"key":57,"meta":731,"navigation":38,"path":732,"problem":733,"proof":745,"recognition":757,"recommendedApproach":772,"relations":774,"seo":779,"stem":782,"summary":783,"visuals":784,"__hash__":785},"useCases_en\u002Fuse-cases\u002Fai-toevoegen-aan-bestaande-software.md","Add AI to existing software, an app or SaaS",{"type":23,"value":714,"toc":715},[],{"title":26,"searchDepth":27,"depth":27,"links":716},[],{"intro":718,"items":719},"The new AI functionality should feel like part of the product, not like a chatbot glued on later.\n",[720,723,726],{"title":721,"text":722},"AI fits the existing user flow","The feature supports what users are already trying to do.\n",{"title":724,"text":725},"The existing architecture stays manageable","AI is integrated without needlessly adding a second technical world.\n",{"title":727,"text":728},"Production risks are covered","Costs, latency, errors and uncertain model behaviour get a place in the design.\n",{"intro":730},"You don't need to build a completely new AI product. Often most of the value lies in one smart feature inside software users already know. We make sure that feature fits the existing product technically.\n",{},"\u002Fuse-cases\u002Fai-toevoegen-aan-bestaande-software",{"intro":734,"causes":735},"Integrating a standalone AI API is easy. Designing AI as a reliable part of existing product software is another story.\n",[736,739,742],{"title":737,"text":738},"Existing software already has rules","Data, permissions, business logic and user expectations are largely fixed.\n",{"title":740,"text":741},"AI introduces other failure modes","Latency, incorrect output and variable costs have to be handled in existing flows.\n",{"title":743,"text":744},"The new feature can end up too detached","A separate AI system quickly leads to duplicated data and logic.\n",{"intro":746,"points":747},"Understanding and extending existing software is a big part of our work. So AI doesn't need a greenfield to be built well.\n",[748,751,754],{"title":749,"text":750},"Understand the existing architecture first","We design the AI feature around what already works.\n",{"title":752,"text":753},"Integration over isolation","Data, auth and business logic are not duplicated unnecessarily.\n",{"title":755,"text":756},"Production from the design onwards","Reliability and operational consequences count before the feature goes live.\n",{"intro":758,"signals":759},"The problem is usually not how to call a model, but how the new functionality should relate to the existing product.\n",[760,763,766,769],{"title":761,"text":762},"The feature needs existing product data","AI has to work with users, documents or other information already in the product.\n",{"title":764,"text":765},"Permissions must simply keep working","AI must not suddenly expose information or actions that are protected elsewhere.\n",{"title":767,"text":768},"The existing UX should stay in charge","Users want a better workflow, not necessarily a chat interface.\n",{"title":770,"text":771},"You don't want to rebuild the current product","The AI layer has to connect to what already works well technically.\n",{"intro":773},"With Build & Integrate we add the AI functionality to the existing codebase and systems in a targeted way. When the new component is technically risky or still uncertain in substance, an AI Prototype Sprint can first prove the core.\n",{"buyerTriggerKeys":775,"primaryOfferKey":65,"offerKeys":776,"caseKeys":777,"relatedUseCaseKeys":778},[35],[65,66],[],[62,58],{"title":780,"description":781},"Add AI to existing software or SaaS | 9to5 software","Add AI features (ChatGPT, Claude, OpenAI API) to existing software? 9to5 designs, builds and integrates AI into existing apps and SaaS products.\n","use-cases\u002Fai-toevoegen-aan-bestaande-software","The product already exists. Now you want to add AI functionality without losing sight of the existing architecture, user experience and reliability.\n",{"hero":217,"problem":75,"proof":96},"uIjq2lxyg3PeXyGhq82YeTX4j_ALcHznMSqDHe95r9k",{"id":787,"title":788,"body":789,"description":26,"desiredOutcome":793,"extension":32,"hero":805,"key":58,"meta":807,"navigation":38,"path":808,"problem":809,"proof":821,"recognition":833,"recommendedApproach":848,"relations":850,"seo":855,"stem":858,"summary":859,"visuals":860,"__hash__":861},"useCases_en\u002Fuse-cases\u002Feigen-data-en-kennis-gebruiken-met-ai.md","Make your own documents, data and knowledge usable for AI",{"type":23,"value":790,"toc":791},[],{"title":26,"searchDepth":27,"depth":27,"links":792},[],{"intro":794,"items":795},"You want to give AI access to relevant information of your own without putting everything into one uncontrollable data bin.\n",[796,799,802],{"title":797,"text":798},"Relevant information at the right moment","The AI only gets the context needed for the current task.\n",{"title":800,"text":801},"Sources stay traceable","Important output can be traced back to the underlying information.\n",{"title":803,"text":804},"Permissions and freshness stay intact","Users only see what they are allowed to see, and changes in sources can carry through into the solution.\n",{"intro":806},"A language model doesn't know your organisation by itself. If AI has to work with your own documents, product data or internal knowledge, that information has to be made available at the right moment and with the right permissions.\n",{},"\u002Fuse-cases\u002Feigen-data-en-kennis-gebruiken-met-ai",{"intro":810,"causes":811},"\"Giving our own data to AI\" sounds like one technical step, but usually consists of selection, retrieval, permissions and quality control.\n",[812,815,818],{"title":813,"text":814},"More context is not automatically better","A model actually gives worse answers when too much irrelevant information is included.\n",{"title":816,"text":817},"Sources have different structures","A PDF, a relational database and a ticketing system each call for a different approach.\n",{"title":819,"text":820},"Data permissions don't disappear with AI","Retrieval has to respect the same access that applies elsewhere in the organisation.\n",{"intro":822,"points":823},"The quality of an AI solution is often determined more by the context you provide than by the newest model. That's why we treat retrieval and data integration as fully-fledged engineering.\n",[824,827,830],{"title":825,"text":826},"Source and use case belong together","We don't just index everything because we can.\n",{"title":828,"text":829},"Existing permissions stay in charge","AI must not become a way around the access model.\n",{"title":831,"text":832},"No unnecessary platform layer","We only choose technical components the real problem requires.\n",{"intro":834,"signals":835},"This question arises when general AI works well, but business-specific context is missing.\n",[836,839,842,845],{"title":837,"text":838},"Knowledge is spread across many sources","Documents, databases and other systems each hold part of the context.\n",{"title":840,"text":841},"Copying into a prompt doesn't scale","Adding information by hand works for a demo, but not for structural use.\n",{"title":843,"text":844},"Information changes regularly","A static dataset quickly becomes outdated.\n",{"title":846,"text":847},"Not all knowledge is for everyone","Access rules from existing systems have to be preserved.\n",{"intro":849},"With Build & Integrate we build the technical layer that gives AI targeted access to your data and knowledge. If feasibility or answer quality is still uncertain, an AI Prototype Sprint can first validate the most important sources and use case.\n",{"buyerTriggerKeys":851,"primaryOfferKey":65,"offerKeys":852,"caseKeys":853,"relatedUseCaseKeys":854},[35],[65,66],[],[55,57],{"title":856,"description":857},"Use your own data and knowledge with AI | 9to5 software","Put your own documents and databases to work with AI? 9to5 builds RAG, retrieval, integrations and AI workflows around your existing information.\n","use-cases\u002Feigen-data-en-kennis-gebruiken-met-ai","The valuable information is already there. Now AI has to be able to work with it reliably, securely and practically.\n",{"hero":631,"problem":217,"proof":96},"rQE5CNfHzt2_sDfHXKyafJGMwiRmtAudJssA1bEegxQ",{"id":863,"title":864,"body":865,"description":26,"desiredOutcome":869,"extension":32,"hero":881,"key":59,"meta":883,"navigation":38,"path":884,"problem":885,"proof":897,"recognition":909,"recommendedApproach":924,"relations":926,"seo":931,"stem":934,"summary":935,"visuals":936,"__hash__":937},"useCases_en\u002Fuse-cases\u002Fdocumenten-en-mails-automatisch-verwerken.md","Process documents, messages, emails or requests automatically",{"type":23,"value":866,"toc":867},[],{"title":26,"searchDepth":27,"depth":27,"links":868},[],{"intro":870,"items":871},"You want to process faster without uncertain AI output automatically ending up in your systems as the truth.\n",[872,875,878],{"title":873,"text":874},"Recognise information automatically","Important fields, topics or intents are extracted from unstructured input.\n",{"title":876,"text":877},"Pass it on automatically where possible","Reliable cases move to the next step without manual retyping.\n",{"title":879,"text":880},"Exceptions go to a person","Doubtful cases are deliberately set aside for review.\n",{"intro":882},"A lot of administrative work starts with the same pattern: something comes in, someone reads it, pulls out the important information and enters it somewhere else. That is where AI and regular automation can work well together.\n",{},"\u002Fuse-cases\u002Fdocumenten-en-mails-automatisch-verwerken",{"intro":886,"causes":887},"Document processing looks like a single AI task at first sight, but a reliable solution consists of several technical steps.\n",[888,891,894],{"title":889,"text":890},"Input varies","Layout, writing style and completeness differ per sender or document.\n",{"title":892,"text":893},"AI output has to be validated","A plausible answer is not automatically correct data for an administrative system.\n",{"title":895,"text":896},"Processing ends somewhere else","The extracted information usually has to go on to a CRM, ERP, tickets or other software.\n",{"intro":898,"points":899},"In document automation, the difference between a demo and a usable system lies in everything around the model output.\n",[900,903,906],{"title":901,"text":902},"Validation is part of the design","Extraction is not blindly treated as correct data.\n",{"title":904,"text":905},"Human-in-the-loop for exceptions","Doubtful cases can be reviewed without making the whole flow manual.\n",{"title":907,"text":908},"The chain gets finished","We connect the output to the systems where the work continues.\n",{"intro":910,"signals":911},"This fits especially when employees spend a lot of time on recurring reading and data entry.\n",[912,915,918,921],{"title":913,"text":914},"A lot of information comes in as free text","Emails, forms and documents don't have a fully fixed structure.\n",{"title":916,"text":917},"Employees keep extracting the same data","Names, amounts, categories, dates or other attributes are copied by hand.\n",{"title":919,"text":920},"A predictable workflow follows","Information is entered, assigned, answered or forwarded.\n",{"title":922,"text":923},"Not every case can be fully automatic","Some documents are incomplete, exceptional or too important for automatic processing.\n",{"intro":925},"With Build & Integrate we build the complete processing flow: input, extraction, validation, human review and connection to existing systems. When document variety or required accuracy is still uncertain, an AI Prototype Sprint can first show what is feasible.\n",{"buyerTriggerKeys":927,"primaryOfferKey":65,"offerKeys":928,"caseKeys":929,"relatedUseCaseKeys":930},[35],[65,66],[],[56,58],{"title":932,"description":933},"Process documents and emails automatically with AI | 9to5 software","Process emails, documents and requests automatically? 9to5 builds AI workflows for extraction, classification, validation and integration with existing systems.\n","use-cases\u002Fdocumenten-en-mails-automatisch-verwerken","Information comes in as free text or documents and then has to be read, assessed, recorded or forwarded.\n",{"hero":631,"problem":347,"proof":708},"ijOxRBH3LNiJuVnuU5z6aIBrVVQEpQfgLLa405EwYiM",{"id":939,"title":940,"body":941,"description":26,"desiredOutcome":945,"extension":32,"hero":956,"key":60,"meta":958,"navigation":38,"path":959,"problem":960,"proof":972,"recognition":984,"recommendedApproach":999,"relations":1001,"seo":1006,"stem":1009,"summary":1010,"visuals":1011,"__hash__":1013},"useCases_en\u002Fuse-cases\u002Fai-prototype-laten-bouwen.md","Build an AI concept or prototype quickly",{"type":23,"value":942,"toc":943},[],{"title":26,"searchDepth":27,"depth":27,"links":944},[],{"intro":946,"items":947},"You don't want slides about what AI could do. You want something that works, so you can make a real technical decision.\n",[948,951,953],{"title":949,"text":950},"A working core flow","The most important user flow works from start to finish.\n",{"title":267,"text":952},"You see whether model, data and integrations work together well enough.\n",{"title":954,"text":955},"A clear follow-up decision","After the prototype you know better what production requires and whether building on makes sense.\n",{"intro":957},"Sometimes you know exactly what you want, but not yet whether it will work well enough technically. Then a full production project is too early. We first build the smallest serious prototype that lets you test the key assumptions.\n",{},"\u002Fuse-cases\u002Fai-prototype-laten-bouwen",{"intro":961,"causes":962},"A prototype should remove uncertainty. If it tries to be the full product already, it becomes unnecessarily slow and expensive.\n",[963,966,969],{"title":964,"text":965},"Scope grows easily","As soon as something works, ideas for accounts, dashboards and extra features appear.\n",{"title":967,"text":968},"Production requirements distract from the core question","The first goal is proving that the most important AI assumption works.\n",{"title":970,"text":971},"Technical feasibility is hard to predict on paper","Some model and data questions you simply have to try.\n",{"intro":973,"points":974},"We build a prototype so you can base a technical decision on it. It is as big as it needs to be to test the uncertainty honestly.\n",[975,978,981],{"title":976,"text":977},"Working instead of theoretical","The most important technical assumption is actually implemented.\n",{"title":979,"text":980},"Scope stays deliberately small","No complete product posing as a prototype.\n",{"title":982,"text":983},"The follow-up is a new choice","Even after a successful sprint you are not obliged to continue.\n",{"intro":985,"signals":986},"A prototype helps most when the biggest uncertainty is technical and won't go away with another meeting.\n",[987,990,993,996],{"title":988,"text":989},"The concept is already fairly sharp","You know which user needs to be able to solve which problem.\n",{"title":991,"text":992},"One technical assumption determines a lot","For example answer quality, document interpretation or agent behaviour.\n",{"title":994,"text":995},"Building in full right away feels too early","There is still too much uncertainty for a larger fixed scope.\n",{"title":997,"text":998},"A real demo would settle the discussion","Stakeholders can decide better when they actually see the system working.\n",{"intro":1000},"The AI Prototype Sprint is the primary route here: one clear use case, one working end-to-end flow and at most the most important integrations. Afterwards you get not only the prototype but also advice on a possible production version.\n",{"buyerTriggerKeys":1002,"primaryOfferKey":66,"offerKeys":1003,"caseKeys":1004,"relatedUseCaseKeys":1005},[35],[66,65],[],[54,57],{"title":1007,"description":1008},"Have an AI prototype built | 9to5 software","Validate an AI concept quickly with a working prototype? 9to5 builds an end-to-end AI prototype that tests the most important technical assumptions.\n","use-cases\u002Fai-prototype-laten-bouwen","The idea is concrete enough to build, but you first want to show that the most important technical assumption actually works.\n",{"hero":75,"problem":1012,"proof":76},"sq-orient-options","2TwZUJkbgDuCmkqkRlk3otdKXvT8Qwxjs8j3pKfpC-M",{"id":1015,"title":1016,"body":1017,"description":26,"desiredOutcome":1021,"extension":32,"hero":1033,"key":61,"meta":1035,"navigation":38,"path":1036,"problem":1037,"proof":1049,"recognition":1061,"recommendedApproach":1076,"relations":1078,"seo":1083,"stem":1086,"summary":1087,"visuals":1088,"__hash__":1089},"useCases_en\u002Fuse-cases\u002Fproduct-goed-opzetten-voor-ai-development.md","Set up a product properly for AI so our own team can then continue independently",{"type":23,"value":1018,"toc":1019},[],{"title":26,"searchDepth":27,"depth":27,"links":1020},[],{"intro":1022,"items":1023},"You want to build as much as possible yourself without early speed later turning into an incomprehensible codebase.\n",[1024,1027,1030],{"title":1025,"text":1026},"A professional start","Stack, architecture, repository and data model together form a clear foundation.\n",{"title":1028,"text":1029},"AI gets good guardrails","Tests, conventions and instructions help coding tools stay within the intended structure.\n",{"title":1031,"text":1032},"The team can carry on itself","Deployment, documentation and way of working are set up for building on independently.\n",{"intro":1034},"AI coding tools can speed you up enormously, especially when they work in a codebase with clear boundaries and feedback. We lay that foundation. After that you don't need to call us for every feature.\n",{},"\u002Fuse-cases\u002Fproduct-goed-opzetten-voor-ai-development",{"intro":1038,"causes":1039},"AI makes producing code cheap. That makes technical direction, feedback and project structure all the more important.\n",[1040,1043,1046],{"title":1041,"text":1042},"The AI optimises for the next task","Without a clear architecture, each local solution needn't fit the bigger whole.\n",{"title":1044,"text":1045},"Errors can multiply quickly","A coding tool easily repeats a bad pattern in several places.\n",{"title":1047,"text":1048},"A good foundation costs much less later","Fixing auth, deployment and project structure is harder once the product is already large.\n",{"intro":1050,"points":1051},"This offer is meant to keep you from depending on an agency for the long term. We do the hard senior work properly and then step aside.\n",[1052,1055,1058],{"title":1053,"text":1054},"Built for AI-assisted development","Structure and feedback help both people and coding agents make better changes.\n",{"title":1056,"text":1057},"One end-to-end path validated","Besides folders and architecture diagrams, we check that the foundation really works.\n",{"title":1059,"text":1060},"Handover is part of the product","The end goal is that you can carry on without day-to-day 9to5 capacity.\n",{"intro":1062,"signals":1063},"This fits when you deliberately want to develop yourself and mainly need senior engineering for the foundation.\n",[1064,1067,1070,1073],{"title":1065,"text":1066},"You want to build with Claude Code, Codex, Cursor, Lovable or Bolt","AI-assisted development becomes an important part of the way you work.\n",{"title":1068,"text":1069},"The product is more serious than a hobby project","Real users, business data or operational dependencies are coming.\n",{"title":1071,"text":1072},"You don't want to need a development agency from day one","Your own team should be able to do as much as possible independently.\n",{"title":1074,"text":1075},"You don't want to need a rescue later","You'd rather get architecture, auth and deployment right up front.\n",{"intro":1077},"With the Vibe Code Kickstart we lay the foundation for one product or codebase professionally: architecture, project structure, data, auth, deployment, tests, AI coding guardrails and documentation. After that you build on yourselves.\n",{"buyerTriggerKeys":1079,"primaryOfferKey":67,"offerKeys":1080,"caseKeys":1081,"relatedUseCaseKeys":1082},[35],[67,174],[],[57,60],{"title":1084,"description":1085},"Set up software properly for AI development | 9to5 software","Building software yourself with Claude Code, Codex, Cursor, Lovable or Bolt? 9to5 sets up architecture, repo, deployment, auth, tests and guardrails professionally.\n","use-cases\u002Fproduct-goed-opzetten-voor-ai-development","You want to build software quickly yourself with AI coding tools, but first lay a professional technical foundation.\n",{"hero":75,"problem":476,"proof":76},"AByEFSpPcWz-73ln6E45vyp8cZhZVEtQHqnwKbXhtag",{"id":1091,"title":1092,"body":1093,"description":26,"desiredOutcome":1097,"extension":32,"hero":1109,"key":62,"meta":1111,"navigation":38,"path":1112,"problem":1113,"proof":1125,"recognition":1137,"recommendedApproach":1152,"relations":1154,"seo":1159,"stem":1162,"summary":1163,"visuals":1164,"__hash__":1165},"useCases_en\u002Fuse-cases\u002Fagentic-workflows-bestaande-software.md","Extend existing software with agentic workflows",{"type":23,"value":1094,"toc":1095},[],{"title":26,"searchDepth":27,"depth":27,"links":1096},[],{"intro":1098,"items":1099},"You want to make an existing workflow smarter without redesigning the rest of the product.\n",[1100,1103,1106],{"title":1101,"text":1102},"Handle more work automatically","Within clear boundaries, the workflow can gather information and take actions itself.\n",{"title":1104,"text":1105},"Existing software stays in charge","The agent uses current data, APIs, auth and business logic.\n",{"title":1107,"text":1108},"Autonomy stays controllable","Critical steps can be validated or put in front of a user.\n",{"intro":1110},"Not every product needs to become an AI-first product. Sometimes a single agentic workflow is exactly enough: an existing process in which software independently performs a few smart steps before a user takes over again.\n",{},"\u002Fuse-cases\u002Fagentic-workflows-bestaande-software",{"intro":1114,"causes":1115},"Adding an agent to existing software is mainly an architecture question: how much freedom does it get and which existing rules stay in charge?\n",[1116,1119,1122],{"title":1117,"text":1118},"The agent has to work within existing domain rules","AI should not re-implement existing business logic in a different way.\n",{"title":1120,"text":1121},"Tool use has to be safe","A wrong answer is different from a wrong write or delete action.\n",{"title":1123,"text":1124},"State has to stay explicit","Long workflows need to know what has already happened and what still has to happen.\n",{"intro":1126,"points":1127},"Agentic workflows sit where AI meets existing software. That is where our experience with software architecture, integrations and AI engineering comes together.\n",[1128,1131,1134],{"title":1129,"text":1130},"Building on what exists","The agent builds on existing systems and rules.\n",{"title":1132,"text":1133},"Autonomy within limits","Tools, permissions and human-in-the-loop are designed deliberately.\n",{"title":1135,"text":1136},"Built for production too","Retries, logging, state and error handling are part of the build.\n",{"intro":1138,"signals":1139},"Agentic workflows pay off most when an existing feature currently takes several manual or half-manual steps.\n",[1140,1143,1146,1149],{"title":1141,"text":1142},"A user has to take many intermediate steps","Fetching, comparing and passing on information takes unnecessary interaction.\n",{"title":1144,"text":1145},"The next step depends on context","A fixed workflow is too rigid, but full human handling is not always necessary.\n",{"title":1147,"text":1148},"The software already has usable APIs","Many of the required actions already exist technically, but are now triggered by hand.\n",{"title":1150,"text":1151},"Full autonomy is not required","The workflow may run on its own until a logical checkpoint or decision moment.\n",{"intro":1153},"With Build & Integrate we design the agentic workflow as an extension of the existing software. We reuse current APIs and domain logic where possible and only build the new orchestration layer that is needed.\n",{"buyerTriggerKeys":1155,"primaryOfferKey":65,"offerKeys":1156,"caseKeys":1157,"relatedUseCaseKeys":1158},[35],[65,66],[],[54,57],{"title":1160,"description":1161},"Add agentic workflows to existing software | 9to5 software","Add agentic workflows or AI agents (Claude, OpenAI) to existing software? 9to5 builds controlled agent flows around your data, APIs and MCP tools.\n","use-cases\u002Fagentic-workflows-bestaande-software","Your existing software already does a lot well, but some workflows could be smarter if AI gathered information and carried out next steps on its own.\n",{"hero":347,"problem":217,"proof":554},"l-LSbU1lEyqddJX5i8k4FjICWCwevjAMP1tFWXQP80s",{"id":1167,"title":1168,"body":1169,"description":26,"desiredOutcome":1173,"extension":32,"hero":1185,"key":63,"meta":1187,"navigation":38,"path":1188,"problem":1189,"proof":1201,"recognition":1213,"recommendedApproach":1228,"relations":1230,"seo":1235,"stem":1238,"summary":1239,"visuals":1240,"__hash__":1241},"useCases_en\u002Fuse-cases\u002Ftools-vervangen-door-maatwerkflow.md","Replace several existing tools with one smart custom workflow",{"type":23,"value":1170,"toc":1171},[],{"title":26,"searchDepth":27,"depth":27,"links":1172},[],{"intro":1174,"items":1175},"You want to simplify the process without blindly rebuilding all existing systems.\n",[1176,1179,1182],{"title":1177,"text":1178},"One clear workflow","Users no longer have to stitch different tools and intermediate steps together themselves.\n",{"title":1180,"text":1181},"Automation where it helps","Manual copying, checking and routing is removed as much as possible.\n",{"title":1183,"text":1184},"Keep what works well","Systems that add value can stay part of the new flow through integrations.\n",{"intro":1186},"Sometimes the problem isn't too little software, but too much. Employees jump between tools, copy information and keep track of how it all fits together themselves. Then one targeted custom workflow can be simpler than yet another SaaS subscription.\n",{},"\u002Fuse-cases\u002Ftools-vervangen-door-maatwerkflow",{"intro":1190,"causes":1191},"Separate tools each optimise their own part of the process. As a result, the complexity often arises precisely between those tools.\n",[1192,1195,1198],{"title":1193,"text":1194},"Integration is missing","Data and status have to be moved by hand from one system to another.\n",{"title":1196,"text":1197},"The process has adapted to the software","Employees follow extra steps because separate tools don't fit together well.\n",{"title":1199,"text":1200},"Otherwise AI becomes yet another tool","A standalone chatbot on top of the existing pile doesn't reduce fragmentation.\n",{"intro":1202,"points":1203},"We don't replace standard SaaS with custom software by default. Because we both build software and connect systems, we can choose per part what is technically and commercially wise.\n",[1204,1207,1210],{"title":1205,"text":1206},"Build and integrate","A good existing system doesn't need to be rebuilt.\n",{"title":1208,"text":1209},"The process leads","The new flow is designed around how the work ideally runs.\n",{"title":1211,"text":1212},"AI only where it fits","Smart interpretation is combined with regular automation and integrations.\n",{"intro":1214,"signals":1215},"This question often arises after a process has been extended bit by bit with new tools over the years.\n",[1216,1219,1222,1225],{"title":1217,"text":1218},"Employees work in many systems for one task","The process keeps jumping between SaaS tools, spreadsheets, email and internal software.\n",{"title":1220,"text":1221},"Information is entered twice","The same data is kept or copied in several places.\n",{"title":1223,"text":1224},"Nobody has an overview of the whole flow","Each tool solves one part, but the employee manually keeps it all together.\n",{"title":1226,"text":1227},"Another new tool doesn't feel like the solution","The pile of software has by now become part of the problem itself.\n",{"intro":1229},"With Build & Integrate we design and build the custom workflow around the actual process. We only replace what makes sense and connect the systems that are better kept. If scope and process are still not sharp enough, an AI Reality Check is the more logical first step.\n",{"buyerTriggerKeys":1231,"primaryOfferKey":65,"offerKeys":1232,"caseKeys":1233,"relatedUseCaseKeys":1234},[35],[65,172],[],[56,62],{"title":1236,"description":1237},"Replace existing tools with a smart custom workflow | 9to5 software","Too many separate tools for one business process? 9to5 builds integrated custom workflows with automation, AI and connections to what should stay.\n","use-cases\u002Ftools-vervangen-door-maatwerkflow","The process has become fragmented across several tools, spreadsheets and manual handoffs. You want to bring it back to one coherent workflow.\n",{"hero":347,"problem":217,"proof":708},"mgSW-1h2BBXZlMWEViIVrX1-qaoPuBUr_YQ6D9XnX9A",[1243,1302],{"id":1244,"title":1245,"approach":1246,"body":1261,"cta":1265,"description":26,"extension":32,"hero":1268,"key":305,"label":1270,"meta":1271,"navigation":38,"path":1272,"proof":1273,"relations":1285,"seo":1294,"stem":1297,"summary":1298,"visuals":1299,"__hash__":1301},"buyerTriggers_en\u002Fbuyer-triggers\u002Fstarten-met-ai.md","We want to get started with AI, but aren't sure how.",{"title":1247,"steps":1248},"From a challenge to a sound technical approach.",[1249,1252,1255,1258],{"title":1250,"text":1251},"Define the problem","We establish which decision you need to make, which process you want to improve and what a good result would look like.\n",{"title":1253,"text":1254},"Understand the context and systems","We look at existing software, data, integrations, users and technical constraints.\n",{"title":1256,"text":1257},"Assess the options","We compare AI, automation, existing software, integrations and custom development in terms of feasibility, risk and value.\n",{"title":1259,"text":1260},"Deliver a decision and a way forward","You get a concrete technical direction, the main risks and a logical next step.\n",{"type":23,"value":1262,"toc":1263},[],{"title":26,"searchDepth":27,"depth":27,"links":1264},[],{"title":1266,"body":1267},"Not sure whether AI is the right solution here?","Briefly describe the idea or process you have in mind. We'll quickly tell you what looks technically promising and what a sensible first step would be.\n",{"intro":1269},"You might have an idea, a process that could work better or simply a sense that AI could do more for you. We help you work out what looks promising, where the risks lie and whether AI is actually the right solution.\n","Start with AI",{},"\u002Fbuyer-triggers\u002Fstarten-met-ai",{"intro":1274,"points":1275},"We don't advise from a fixed preference for AI or custom software. Because we build software ourselves, we can honestly assess what is technically feasible and worthwhile, and what you are better off not building.\n",[1276,1279,1282],{"title":1277,"text":1278},"Engineers look at the real system","We look at the AI idea, but also at the data, existing software, integrations and users.\n",{"title":1280,"text":1281},"AI is a tool, not the default answer","Automation, existing software, an integration or building nothing are all serious options too.\n",{"title":1283,"text":1284},"Advice and implementation from the same people","If building turns out to make sense, we can carry out the next steps ourselves. That keeps our advice realistic.\n",{"useCaseKeys":1286,"offerKeys":1292,"caseKeys":1293},[1287,1288,1289,1290,244,1291,245,374],"processen-automatiseren-met-ai","interne-ai-assistent-met-eigen-kennis","ai-idee-technisch-haalbaar","is-ai-nodig","zelf-bouwen-of-software-kopen",[172,67,66],[],{"title":1295,"description":1296},"Getting started with AI? From idea to technical approach | 9to5 software","Help with a serious AI challenge. An AI agent, ChatGPT, Claude or conventional software? We assess what is feasible, what not to build and the next step.\n","buyer-triggers\u002Fstarten-met-ai","We see opportunities for AI, but first want to know what makes sense, what is feasible and what is technically sound.\n",{"hero":1012,"proof":1300,"approach":1300},"ls-route-converge","oMS04pD9lTBstiI46yn57yKInU-_urtigoQYGlj0PWA",{"id":1303,"title":1304,"approach":1305,"body":1320,"cta":1324,"description":26,"extension":32,"hero":1327,"key":169,"label":1329,"meta":1330,"navigation":38,"path":1331,"proof":1332,"relations":1344,"seo":1366,"stem":1369,"summary":1370,"visuals":1371,"__hash__":1374},"buyerTriggers_en\u002Fbuyer-triggers\u002Fvastgelopen-met-ai.md","We've already built something, but we're stuck.",{"title":1306,"steps":1307},"We step in, understand, repair and hand over.",[1308,1311,1314,1317],{"title":1309,"text":1310},"Confirm the rescue goal","We establish what is actually going wrong, what is urgent and what needs to be demonstrably better after the rescue.\n",{"title":1312,"text":1313},"Understand the context and code","We examine the code, architecture, data flows, integrations and deployment until the source of the problem is clear.\n",{"title":1315,"text":1316},"Fix and test","We make targeted repairs, add tests, evals and guardrails where needed and check that the solution works under real conditions too.\n",{"title":1318,"text":1319},"Document and hand over","We document decisions, clear critical technical debt and make sure your team can move forward independently again.\n",{"type":23,"value":1321,"toc":1322},[],{"title":26,"searchDepth":27,"depth":27,"links":1323},[],{"title":1325,"body":1326},"Stuck with AI or vibe-coded software?","Briefly describe what you have and where it is going wrong. We'll take a look and quickly tell you what makes sense to repair, keep or rebuild.\n",{"intro":1328},"You don't have to start over just because things have become complicated. We step in, including when you have a vibe-coded app or code generated by AI tools such as Claude Code, Codex or Cursor. We understand what is there, find what is really going wrong and make it technically manageable again.\n","Stuck with AI",{},"\u002Fbuyer-triggers\u002Fvastgelopen-met-ai",{"intro":1333,"points":1334},"Assessing an existing codebase takes more than an automated quality score. It takes senior engineers who can identify the problems that really hurt and make targeted repairs.\n",[1335,1338,1341],{"title":1336,"text":1337},"We work with existing software","A messy or AI-generated codebase is no reason to automatically start from scratch.\n",{"title":1339,"text":1340},"Software engineering and AI","We look at everyday production problems and AI-specific failure modes, such as unreliable output and missing evals.\n",{"title":1342,"text":1343},"Move forward independently again","We document decisions and clear critical technical debt so your team can move forward on its own again.\n",{"useCaseKeys":1345,"offerKeys":1364,"caseKeys":1365},[1346,1347,1348,1349,1350,1351,1352,1353,1354,1355,1356,1357,1358,1359,1360,1361,1362,1363],"vibe-coded-app-professionaliseren","ai-prototype-naar-productie","ai-agent-niet-betrouwbaar","gegenereerde-code-niet-meer-begrijpelijk","ai-code-breekt-steeds","vast-op-auth-database-of-rechten","ai-koppelen-aan-bestaande-systemen","security-en-privacy-ai-oplossing","ai-oplossing-te-langzaam","ai-oplossing-te-duur","tests-evals-monitoring-voor-ai","developmentteam-loopt-vast-met-ai","technische-second-opinion-ai","ai-project-overnemen-van-andere-leverancier","ai-experiment-heeft-echte-gebruikers","prototype-naar-onderhoudbaar-softwareproduct","weg-bij-ai-platform-of-leverancier","vibe-code-project-vastgelopen",[173,174,65],[],{"title":1367,"description":1368},"Stuck with AI? Rescue by senior engineers | 9to5 software","Stuck with an AI agent, prototype or vibe-coded app (Lovable, Cursor, Bolt, Replit)? We analyse the problem, make targeted repairs and restore maintainability.\n","buyer-triggers\u002Fvastgelopen-met-ai","Something is already there, but it is unreliable, hard to understand or technically stuck. We need to get it back under control.\n",{"hero":1372,"proof":1373,"approach":1373},"sq-tangle-knot","ls-rescue-repair","e0bDu3r1P4JqQOadOsQgPsL_EDOsQGpgsqBXrdwFTNw",[],{"id":1377,"title":1378,"extension":1379,"items":1380,"meta":1393,"stem":1394,"__hash__":1395},"why9to5_en\u002Fen\u002Fshared\u002Fwhy-9to5.yml","Why 9to5?","yml",[1381,1384,1387,1390],{"label":1382,"value":1383},"Senior engineering","AI + software engineering",{"label":1385,"value":1386},"Existing software","We step in where it gets complicated",{"label":1388,"value":1389},"Production-ready","Built, tested and reliably deployed",{"label":1391,"value":1392},"Handover","Documented, so you can carry on independently",{},"en\u002Fshared\u002Fwhy-9to5","_iz4GBKRty2Gu5JZKf6tSCRXdwX0sju7lRprYEzQvNE",1792049913948]