[{"data":1,"prerenderedAt":936},["ShallowReactive",2],{"insight-insights_en-build-buy-or-both":3,"insight-related-insights_en-build-buy-or-both":278},{"id":4,"title":5,"author":6,"blobHue":10,"body":11,"category":253,"date":254,"description":17,"draft":255,"extension":256,"featured":255,"headline":257,"hue":258,"letter":259,"meta":260,"navigation":261,"path":262,"readMinutes":263,"related":264,"seo":268,"stem":269,"summary":270,"toc":271,"__hash__":277},"insights_en\u002Finsights\u002Fbuild-buy-or-both.md","Build, buy or both?",{"name":7,"role":8,"bio":9},"André","Founder & CTO","André is the founder and CTO of WizardingCode. Eight years building the software companies run on, now putting agents into production.",null,{"type":12,"value":13,"toc":244},"minimark",[14,18,21,26,29,37,44,50,53,57,60,88,91,95,98,104,110,196,199,205,209,212,215,219,241],[15,16,17],"p",{},"Every company we talk to already pays for some AI. A writing assistant here, a meeting summariser there, a chatbot on the website. The question is rarely whether to use AI. It is which work deserves more than a subscription.",[15,19,20],{},"The honest answer is that most of it doesn’t. And the part that does usually needs both: things you buy and things you build.",[22,23,25],"h2",{"id":24},"when-buying-is-enough","When buying is enough",[15,27,28],{},"An off-the-shelf tool is the right choice when three things are true.",[15,30,31,32,36],{},"The work is ",[33,34,35],"strong",{},"generic",". Summarising a meeting, rewriting an email, translating a document, drafting a first version of a job ad. Every company does it roughly the same way, so a product built for everyone fits you well.",[15,38,39,40,43],{},"It needs ",[33,41,42],{},"no live data",". The input is whatever the person pastes in, and the output goes back to that person. Nothing has to be read from your ERP or written into your CRM.",[15,45,46,49],{},[33,47,48],{},"Nobody owns an outcome."," It makes individuals faster, but no team has signed up to move a number with it. If it gets worse next month, someone switches tools and nothing breaks.",[15,51,52],{},"For this kind of work, building your own is a waste of money. Buy the tool, set some sensible rules on what data goes into it, and move on.",[22,54,56],{"id":55},"when-a-process-deserves-its-own-agent","When a process deserves its own agent",[15,58,59],{},"The picture changes when the work runs through your systems and your rules. A process deserves its own agent when:",[61,62,63,70,76,82],"ol",{},[64,65,66,69],"li",{},[33,67,68],{},"It lives in your systems."," The agent has to read the order, the ticket history, the supplier file or the price list, and write back to them.",[64,71,72,75],{},[33,73,74],{},"There is a number."," First response time, hours spent on admin, stock left unsold. A team owns it and wants it to move.",[64,77,78,81],{},[33,79,80],{},"Some decisions need a person."," Refunds, promises, anything that cannot be undone. Those need approval rules that match how your company works, not how a vendor imagined companies work.",[64,83,84,87],{},[33,85,86],{},"It is specific to you."," Your policies, your exceptions, your tone, the thing only Maria knows. A generic tool does not know any of it and you cannot teach it properly.",[15,89,90],{},"No product sold to everyone can do all four for you. It doesn’t know your systems, it can’t own your number, and its rules are someone else’s.",[22,92,94],{"id":93},"the-both-pattern","The both pattern",[15,96,97],{},"In practice, the answer for most serious processes is both.",[15,99,100,103],{},[33,101,102],{},"You buy the commodity."," The language models themselves, from the big model providers. Hosting, email delivery, search, the plumbing. Nobody should build their own model to answer supplier emails.",[15,105,106,109],{},[33,107,108],{},"You build what is yours."," The agents, the approval rules and the shared memory. That is where your process, your policies and your judgement live, and it is the part that makes the difference between a demo and a system you can rely on.",[111,112,113,128],"table",{},[114,115,116],"thead",{},[117,118,119,122,125],"tr",{},[120,121],"th",{},[120,123,124],{},"BUY",[120,126,127],{},"BUILD",[129,130,131,145,157,170,183],"tbody",{},[117,132,133,139,142],{},[134,135,136],"td",{},[33,137,138],{},"Models",[134,140,141],{},"From the model providers",[134,143,144],{},"Never",[117,146,147,152,155],{},[134,148,149],{},[33,150,151],{},"Infrastructure",[134,153,154],{},"Hosting, email, storage",[134,156,144],{},[117,158,159,164,167],{},[134,160,161],{},[33,162,163],{},"Agents",[134,165,166],{},"Rarely a fit",[134,168,169],{},"Built around your process",[117,171,172,177,180],{},[134,173,174],{},[33,175,176],{},"Approval rules",[134,178,179],{},"Generic settings",[134,181,182],{},"Written with your team",[117,184,185,190,193],{},[134,186,187],{},[33,188,189],{},"Shared memory",[134,191,192],{},"Empty until you fill it",[134,194,195],{},"Your policies, history and decisions",[15,197,198],{},"This split also protects you. Models improve and prices change every few months. When the agents, rules and memory are yours, swapping the model underneath is an engineering task, not a migration. The marketplace system we built uses models from several providers at once, each chosen for the job, and the system does not depend on any single one.",[200,201,202],"blockquote",{},[15,203,204],{},"Buy the model. Build the part that knows your business.",[22,206,208],{"id":207},"who-owns-what","Who owns what",[15,210,211],{},"Whatever you build, ask who owns it at the end. In our projects the answer is simple. The agents run in your environment and the memory is in your systems. You own the code, the prompts and the data, so you can run them without us.",[15,213,214],{},"That matters more than it seems. A process that runs on an agent you do not own is a process someone else can reprice, change or switch off.",[22,216,218],{"id":217},"a-quick-test","A quick test",[220,221,223],"prose-checklist",{"title":222},"Before you buy another AI tool, ask",[224,225,226,229,232,235,238],"ul",{},[64,227,228],{},"Does it need to read or write your live systems?",[64,230,231],{},"Is there a number that a team is expected to move with it?",[64,233,234],{},"Are there decisions in it that must stay with a person?",[64,236,237],{},"Does it depend on policies, exceptions or tone that are specific to you?",[64,239,240],{},"If the vendor changed its price or product tomorrow, would a process stop?",[15,242,243],{},"If every answer is no, buy it. If two or more are yes, the process probably deserves its own agent, built on bought models, with rules and memory that belong to you.",{"title":245,"searchDepth":246,"depth":246,"links":247},"",2,[248,249,250,251,252],{"id":24,"depth":246,"text":25},{"id":55,"depth":246,"text":56},{"id":93,"depth":246,"text":94},{"id":207,"depth":246,"text":208},{"id":217,"depth":246,"text":218},"strategy","2026-08-04",false,"md","Build, buy or ==both?==","violet","?",{},true,"\u002Finsights\u002Fbuild-buy-or-both",4,[265,266,267],"what-an-agentic-os-is-and-what-it-isnt","why-95-percent-of-ai-pilots-never-reach-production","the-30-day-playbook-week-by-week",{"title":5,"description":17},"insights\u002Fbuild-buy-or-both","When an off-the-shelf AI tool is enough, and when your process deserves its own agent.",[272,273,274,275,276],{"id":24,"label":25},{"id":55,"label":56},{"id":93,"label":94},{"id":207,"label":208},{"id":217,"label":218},"XoWrr_D4QkTedkkrnrgccPyPPGBK8-ZFnFJfkxogvLc",[279,516,726],{"id":280,"title":281,"author":282,"blobHue":10,"body":283,"category":253,"date":498,"description":287,"draft":255,"extension":256,"featured":255,"headline":499,"hue":258,"letter":500,"meta":501,"navigation":261,"path":502,"readMinutes":263,"related":503,"seo":506,"stem":507,"summary":508,"toc":509,"__hash__":515},"insights_en\u002Finsights\u002Fwhat-an-agentic-os-is-and-what-it-isnt.md","What an Agentic OS is, and what it isn’t",{"name":7,"role":8,"bio":9},{"type":12,"value":284,"toc":491},[285,288,291,295,298,301,305,310,342,347,352,358,424,429,433,439,445,451,455,458,461,465,488],[15,286,287],{},"Most people meet AI agents as a chat window. You ask, it answers, and then nothing happens. That is useful, but it isn’t how a business runs. A business runs on work that moves between systems, people and decisions, every day, whether anyone is watching or not.",[15,289,290],{},"An Agentic OS is what it takes to hand some of that work to agents and still sleep at night.",[22,292,294],{"id":293},"a-plain-definition","A plain definition",[15,296,297],{},"An Agentic OS is a team of AI agents trained on how your business works, running inside the tools you already use, following rules your people set. Each agent owns one job. They share what they know. They stop and ask when a decision is above their limits. And everything they do is visible on one screen.",[15,299,300],{},"The word “OS” is deliberate. An operating system is not an app you open. It is what sits underneath and keeps everything else running. The agents are the visible part; the layers around them are what make it safe.",[22,302,304],{"id":303},"the-four-layers","The four layers",[15,306,307,309],{},[33,308,163],{}," do the work. We use five kinds, and most companies go live with two or three:",[61,311,312,318,324,330,336],{},[64,313,314,317],{},[33,315,316],{},"Responder"," answers customers, suppliers and colleagues, in your tone, from your data.",[64,319,320,323],{},[33,321,322],{},"Classifier"," reads what comes in, from emails to tickets to documents, and sends it to the right place.",[64,325,326,329],{},[33,327,328],{},"Scraper"," watches the sources you care about and brings back what changed.",[64,331,332,335],{},[33,333,334],{},"Orchestrator"," runs work that spans several systems: the refund, the CRM update, the courier booking.",[64,337,338,341],{},[33,339,340],{},"Analyst"," reads the numbers and has the report ready before the meeting.",[15,343,344,346],{},[33,345,189],{}," is what the agents know about your business: policies, price lists, customer history, past decisions. Every agent reads from the same source, and it stays current as you work. No more “ask Maria, she knows”.",[15,348,349,351],{},[33,350,176],{}," decide what an agent may do alone. They are written in plain language and set by your team. Above the limit, the agent pauses and a person decides, in the tools they already use.",[15,353,354,357],{},[33,355,356],{},"The command center"," shows every run, cost and result on one screen. You can replay any decision step by step, pause any agent, and follow the number you agreed on, every day.",[111,359,360,373],{},[114,361,362],{},[117,363,364,367,370],{},[120,365,366],{},"LAYER",[120,368,369],{},"WHAT IT ANSWERS",[120,371,372],{},"WITHOUT IT",[129,374,375,387,399,411],{},[117,376,377,381,384],{},[134,378,379],{},[33,380,163],{},[134,382,383],{},"Who does the work?",[134,385,386],{},"Nothing gets done",[117,388,389,393,396],{},[134,390,391],{},[33,392,189],{},[134,394,395],{},"What do they know?",[134,397,398],{},"Every agent guesses",[117,400,401,405,408],{},[134,402,403],{},[33,404,176],{},[134,406,407],{},"What can they do alone?",[134,409,410],{},"Nobody dares switch them on",[117,412,413,418,421],{},[134,414,415],{},[33,416,417],{},"Command center",[134,419,420],{},"What did they do, and did it work?",[134,422,423],{},"Nobody can prove it",[200,425,426],{},[15,427,428],{},"Agents alone are a demo. What makes them safe is everything around them.",[22,430,432],{"id":431},"what-it-isnt","What it isn’t",[15,434,435,438],{},[33,436,437],{},"It isn’t a chatbot."," A chatbot waits for a question. An Agentic OS works through a queue. It reads the inbox overnight, reconciles the stock file, drafts the replies, and leaves the three decisions that need a person at the top of the list in the morning.",[15,440,441,444],{},[33,442,443],{},"It isn’t a platform licence."," You don’t rent seats in someone else’s product and bend your process to fit it. The agents are built around your process, inside your systems, and you own the code, the prompts and the data.",[15,446,447,450],{},[33,448,449],{},"It isn’t a pilot."," A pilot runs on an export, for a demo, with nobody accountable for the result. An Agentic OS runs on live systems from the first week, against a number that the team whose work it takes has signed off. If it isn’t in production, it isn’t an OS yet.",[22,452,454],{"id":453},"where-to-start","Where to start",[15,456,457],{},"Nobody builds all four layers for the whole company at once. You start with one process that hurts, usually one with volume, clear rules and a number: the support inbox, the supplier files, the weekly report. You build the first squad of agents with its memory, its rules and its screen. Then you add a squad at a time.",[15,459,460],{},"The layers are what make the second squad cheaper than the first. The memory is already there. The rules have a format the team knows. The command center already shows the first squad’s numbers, so the next one is judged on the same screen, against the same kind of target.",[22,462,464],{"id":463},"how-to-tell-if-you-have-one","How to tell if you have one",[220,466,468],{"title":467},"You have an Agentic OS if the answer is yes to all of these",[224,469,470,473,476,479,482,485],{},[64,471,472],{},"Do the agents work inside the systems your team already uses?",[64,474,475],{},"Can you say, in one sentence, which decisions they may not take alone?",[64,477,478],{},"Do they all read from the same, current source of policies and history?",[64,480,481],{},"Can you open one screen and see what they did today, what it cost and what is waiting for you?",[64,483,484],{},"Has a number moved, and does the team that owns it agree?",[64,486,487],{},"Do you own the code, the prompts and the data?",[15,489,490],{},"If any answer is no, you have agents. That is a good start. The rest is the part that lets you rely on them.",{"title":245,"searchDepth":246,"depth":246,"links":492},[493,494,495,496,497],{"id":293,"depth":246,"text":294},{"id":303,"depth":246,"text":304},{"id":431,"depth":246,"text":432},{"id":453,"depth":246,"text":454},{"id":463,"depth":246,"text":464},"2026-09-15","What an Agentic OS is, and what it ==isn’t.==","OS",{},"\u002Finsights\u002Fwhat-an-agentic-os-is-and-what-it-isnt",[266,504,505],"every-agent-needs-a-judge","what-agents-should-never-do-alone",{"title":281,"description":287},"insights\u002Fwhat-an-agentic-os-is-and-what-it-isnt","Not a chatbot, not a platform licence. A plain-language definition, with the four layers that make agents safe to run a business.",[510,511,512,513,514],{"id":293,"label":294},{"id":303,"label":304},{"id":431,"label":432},{"id":453,"label":454},{"id":463,"label":464},"O4B58r8Hq-cONZEYpn0ps8PFaK7Sr4eN71hJvwSQj0c",{"id":517,"title":518,"author":519,"blobHue":520,"body":521,"category":253,"date":707,"description":708,"draft":255,"extension":256,"featured":261,"headline":709,"hue":710,"letter":711,"meta":712,"navigation":261,"path":713,"readMinutes":714,"related":715,"seo":716,"stem":717,"summary":718,"toc":719,"__hash__":725},"insights_en\u002Finsights\u002Fwhy-95-percent-of-ai-pilots-never-reach-production.md","Why 95% of AI pilots never reach production",{"name":7,"role":8,"bio":9},"blue",{"type":12,"value":522,"toc":700},[523,531,534,538,543,548,552,555,575,579,582,663,667,687,691,694],[15,524,525,526,530],{},"In 2025, MIT’s NANDA initiative looked at how companies were using generative AI and found that about 95% of enterprise pilots delivered no measurable impact on the P&L.",[527,528,529],"sup",{},"1"," Billions spent, demos applauded, and almost nothing changed in how the business actually ran.",[15,532,533],{},"We’ve seen the same pattern from the inside. Most of the companies that call us have already run a pilot. It worked in the demo. It never went live.",[22,535,537],{"id":536},"the-number-everyone-quotes","The number everyone quotes",[15,539,540,541],{},"The headline is easy to misread. It doesn’t say AI doesn’t work. It says pilots don’t turn into production. The same report found that projects built with specialised external partners reached deployment about twice as often as internal builds, and that the biggest returns came from unglamorous back-office work, not from customer-facing chatbots.",[527,542,529],{},[200,544,545],{},[15,546,547],{},"The model is rarely the problem. The plumbing is.",[22,549,551],{"id":550},"its-not-the-model","It’s not the model",[15,553,554],{},"When we look at why a pilot stalled, the answer is almost never “the AI wasn’t smart enough”. It is one of three things:",[61,556,557,563,569],{},[64,558,559,562],{},[33,560,561],{},"No live data."," The pilot ran on an export. Connecting it to the real ERP, inbox or helpdesk was “phase two”, and phase two never came.",[64,564,565,568],{},[33,566,567],{},"No owner."," The innovation team built it. The team whose work it would take never asked for it, and never adopted it.",[64,570,571,574],{},[33,572,573],{},"No number."," Success was “a good demo”. Nobody agreed what had to move, so nobody could say it had worked.",[22,576,578],{"id":577},"four-things-the-5-do","Four things the 5% do",[15,580,581],{},"The projects that ship look different from day one. They connect to live systems in week one, not month six. The metric is agreed before a line of code is written, and it belongs to the team whose work changes. People approve the decisions that carry risk, so nobody has to trust the agent blindly. And there is a date: production in weeks, not a roadmap.",[111,583,584,596],{},[114,585,586],{},[117,587,588,590,593],{},[120,589],{},[120,591,592],{},"A PILOT",[120,594,595],{},"A SYSTEM IN PRODUCTION",[129,597,598,611,624,637,650],{},[117,599,600,605,608],{},[134,601,602],{},[33,603,604],{},"Data",[134,606,607],{},"A sample export",[134,609,610],{},"Your live systems",[117,612,613,618,621],{},[134,614,615],{},[33,616,617],{},"Success",[134,619,620],{},"A good demo",[134,622,623],{},"A number agreed on day one",[117,625,626,631,634],{},[134,627,628],{},[33,629,630],{},"Owner",[134,632,633],{},"The innovation team",[134,635,636],{},"The team whose work it takes",[117,638,639,644,647],{},[134,640,641],{},[33,642,643],{},"Humans",[134,645,646],{},"Watching",[134,648,649],{},"Approving what matters",[117,651,652,657,660],{},[134,653,654],{},[33,655,656],{},"Timeline",[134,658,659],{},"Open-ended",[134,661,662],{},"Weeks",[22,664,666],{"id":665},"a-checklist-before-you-start","A checklist before you start",[220,668,670],{"title":669},"Before you approve the next AI project, ask",[224,671,672,675,678,681,684],{},[64,673,674],{},"Which live system does it read and write, in week one?",[64,676,677],{},"Which number moves, and who owns it?",[64,679,680],{},"Which decisions stay with a person?",[64,682,683],{},"What happens on day 30?",[64,685,686],{},"Who owns the code, the prompts and the data?",[22,688,690],{"id":689},"what-this-means-for-you","What this means for you",[15,692,693],{},"If your last pilot is gathering dust, it probably wasn’t a bad idea. It was missing the boring parts. Start from one process that hurts, connect it to the real systems, agree the number and put a date on it. That’s the whole method. It isn’t magic; it’s engineering.",[695,696,697],"prose-footnotes",{},[15,698,699],{},"¹ MIT NANDA, “The GenAI Divide: State of AI in Business 2025”.",{"title":245,"searchDepth":246,"depth":246,"links":701},[702,703,704,705,706],{"id":536,"depth":246,"text":537},{"id":550,"depth":246,"text":551},{"id":577,"depth":246,"text":578},{"id":665,"depth":246,"text":666},{"id":689,"depth":246,"text":690},"2026-09-22","In 2025, MIT’s NANDA initiative looked at how companies were using generative AI and found that about 95% of enterprise pilots delivered no measurable impact on the P&L.1 Billions spent, demos applauded, and almost nothing changed in how the business actually ran.","Why 95% of AI pilots never reach ==production.==","magenta","95",{},"\u002Finsights\u002Fwhy-95-percent-of-ai-pilots-never-reach-production",7,[265,504,267],{"title":518,"description":708},"insights\u002Fwhy-95-percent-of-ai-pilots-never-reach-production","It’s rarely the model. It’s the data, the owner and the missing number. Here’s what the ones that ship do differently.",[720,721,722,723,724],{"id":536,"label":537},{"id":550,"label":551},{"id":577,"label":578},{"id":665,"label":666},{"id":689,"label":690},"9wJDwNO8uXgEwGK-cE1WYthNEN-J5Ic1e7kI_crM30g",{"id":727,"title":728,"author":729,"blobHue":10,"body":730,"category":917,"date":918,"description":734,"draft":255,"extension":256,"featured":255,"headline":919,"hue":710,"letter":920,"meta":921,"navigation":261,"path":922,"readMinutes":263,"related":923,"seo":925,"stem":926,"summary":927,"toc":928,"__hash__":935},"insights_en\u002Finsights\u002Fthe-30-day-playbook-week-by-week.md","The 30-day playbook, week by week",{"name":7,"role":8,"bio":9},{"type":12,"value":731,"toc":909},[732,735,738,742,745,748,752,755,781,784,788,791,794,797,801,804,807,889,893,896,899,903,906],[15,733,734],{},"Thirty days sounds fast for putting AI agents into production. It is fast. It is also the reason most of our projects ship: a fixed date forces every decision that a pilot would postpone to be taken in the first week.",[15,736,737],{},"This is the playbook we follow. Every step ends with something written down and someone’s name next to it.",[22,739,741],{"id":740},"week-0-diagnose","Week 0: diagnose",[15,743,744],{},"It starts with a free call. If there is a fit, we spend the next week inside your operations, most of it with the people who do the work. We sit in on the inbox, the report, the supplier file. We want to see where the hours go and where the money leaks, not how the process is described in a slide.",[15,746,747],{},"The artefact is a map of your operations with the value of every agent we would build, and a fixed price. The sign-off is yours: which process goes first. We push for one with volume, clear rules and a number someone already cares about.",[22,749,751],{"id":750},"week-1-map","Week 1: map",[15,753,754],{},"This is the week most pilots skip, and the week that decides whether yours ships.",[61,756,757,763,769,775],{},[64,758,759,762],{},[33,760,761],{},"Access."," Real credentials to the real systems: the CRM, the helpdesk, the ERP, the inbox. Not an export.",[64,764,765,768],{},[33,766,767],{},"Real cases."," A set of past emails, tickets or files, with what a good answer looked like. They become the tests.",[64,770,771,774],{},[33,772,773],{},"One metric, signed off."," First response time, time to spot a new listing, partner hours on admin. One number, owned by the team whose work changes, signed by the person who owns it.",[64,776,777,780],{},[33,778,779],{},"First approval rules."," What the agents may do alone, and what always goes to a person, written in plain language.",[15,782,783],{},"If week 1 ends without access and a signed metric, we say so, because it is cheaper for both sides than finding out on day 29.",[22,785,787],{"id":786},"week-2-build","Week 2: build",[15,789,790],{},"Now we build the agents, the rules, the shared memory and the command center, wired into your tools. The agents live where your team already works; nobody gets a new app to learn.",[15,792,793],{},"The real cases from week 1 become evals: automated tests that run every answer against what a good answer looked like. Customer-facing answers get a judge. Tone is agreed with the people who own it, often in a single working session with a pile of past replies.",[15,795,796],{},"The sign-off at the end of the week is a walkthrough with the team lead, on their own cases.",[22,798,800],{"id":799},"week-3-prove","Week 3: prove",[15,802,803],{},"First the evals, on real cases the agents have never seen. Then shadow mode on live work: the agents draft, people send. Every edit a person makes is a signal, and we read them daily.",[15,805,806],{},"This is also where the handoff rules are tuned with the team. Which cases go straight to a person, at what threshold, through which channel. The team should be able to change a rule themselves, and in week 3 they practise doing it.",[111,808,809,822],{},[114,810,811],{},[117,812,813,816,819],{},[120,814,815],{},"STEP",[120,817,818],{},"ARTEFACT",[120,820,821],{},"WHO SIGNS",[129,823,824,837,850,863,876],{},[117,825,826,831,834],{},[134,827,828],{},[33,829,830],{},"Week 0",[134,832,833],{},"Operations map, value per agent, fixed price",[134,835,836],{},"You: which process first",[117,838,839,844,847],{},[134,840,841],{},[33,842,843],{},"Week 1",[134,845,846],{},"Access, real cases, first approval rules",[134,848,849],{},"The owner of the metric",[117,851,852,857,860],{},[134,853,854],{},[33,855,856],{},"Week 2",[134,858,859],{},"Agents, memory, rules, command center, evals",[134,861,862],{},"The team lead, after a walkthrough",[117,864,865,870,873],{},[134,866,867],{},[33,868,869],{},"Week 3",[134,871,872],{},"Eval results, shadow-mode log, tuned handoffs",[134,874,875],{},"The team lead: ready to go live",[117,877,878,883,886],{},[134,879,880],{},[33,881,882],{},"Day 30",[134,884,885],{},"Production, monitoring, trained team",[134,887,888],{},"Both of us, against the metric",[22,890,892],{"id":891},"day-30-live","Day 30: live",[15,894,895],{},"On day 30 the agents are in production, on live work, monitored from the command center. The team is trained on it: how to read a run, how to approve, how to pause an agent, how to change a rule.",[15,897,898],{},"Production means the agents answer, route or update on their own inside the rules, and the number from week 1 is tracked every day from then on. It does not mean “available on request” or “ready for phase two”.",[22,900,902],{"id":901},"why-the-date-holds","Why the date holds",[15,904,905],{},"Three things keep the date honest. The scope is one process, not the company. The metric is agreed before a line of code, so nobody can move the goalposts later. And we put our money on it: live in 30 days, or your money back.",[15,907,908],{},"After day 30 we run the fleet: monitoring, tuning and, when you are ready, the next squad. On our largest project that became one new squad a month, each built on the memory and rules of the ones before it.",{"title":245,"searchDepth":246,"depth":246,"links":910},[911,912,913,914,915,916],{"id":740,"depth":246,"text":741},{"id":750,"depth":246,"text":751},{"id":786,"depth":246,"text":787},{"id":799,"depth":246,"text":800},{"id":891,"depth":246,"text":892},{"id":901,"depth":246,"text":902},"playbooks","2026-08-18","The 30-day playbook, week by ==week.==","30",{},"\u002Finsights\u002Fthe-30-day-playbook-week-by-week",[266,505,924],"build-buy-or-both",{"title":728,"description":734},"insights\u002Fthe-30-day-playbook-week-by-week","From the first call to production: the exact steps, artefacts and sign-offs we use.",[929,930,931,932,933,934],{"id":740,"label":741},{"id":750,"label":751},{"id":786,"label":787},{"id":799,"label":800},{"id":891,"label":892},{"id":901,"label":902},"rT6HBQ-x40ciVZH-qHpcDH1wLuGv79MQR2E3ffHF_dY",1790618465055]