<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Maxime Lenne on Medium]]></title>
        <description><![CDATA[Stories by Maxime Lenne on Medium]]></description>
        <link>https://medium.com/@MaximeLenne?source=rss-7ea898cb1bf4------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*ldYNTpCYliel13fK.jpeg</url>
            <title>Stories by Maxime Lenne on Medium</title>
            <link>https://medium.com/@MaximeLenne?source=rss-7ea898cb1bf4------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 09:12:31 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@MaximeLenne/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Nord Star Product 2026, Aurora edition: my takeaways from the 7 talks]]></title>
            <link>https://medium.com/@MaximeLenne/nord-star-product-2026-aurora-edition-my-takeaways-from-the-7-talks-53bbe737252d?source=rss-7ea898cb1bf4------2</link>
            <guid isPermaLink="false">https://medium.com/p/53bbe737252d</guid>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[design-sprint]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[ux-research]]></category>
            <category><![CDATA[double-diamond]]></category>
            <dc:creator><![CDATA[Maxime Lenne]]></dc:creator>
            <pubDate>Sat, 24 Oct 2026 00:00:35 GMT</pubDate>
            <atom:updated>2026-10-30T08:41:01.646Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LLF6u-5LdTn-mHnwUnUr4g.jpeg" /></figure><p><strong>“We didn’t save time to go faster, but to spend more time on what really matters.”</strong> Anthony Lefebvre and Mathéo Evrard said it on the Nord Star Product stage on 22 October 2026, and it sums up the day quite well. Everyone talked about AI. Almost nobody said it would make decisions for us.</p><p>Here is my recap of the Aurora edition: the programme, the ideas and quotes from each talk, then what I take away from it, on the tech side and the human side. The talks were in French; the quotes below are my translations.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cYaUUO61fwpiF4Ud3lovfQ.jpeg" /></figure><h3>Nord Star Product, Aurora edition</h3><p><a href="https://nordstarproduct.com/lineup">Nord Star Product</a> is the design and product day in Lille. It is organised by the Nord Star Product community and EuraCreative, which supports cultural and creative companies on the Plaine Images campus.</p><ul><li>Date: Thursday 22 October 2026.</li><li>Venue: Plaine Images, Tourcoing, near Lille.</li><li>Format: 45 minutes of talk in the room, then a 15-minute break outside, because “the best moments of a conference happen at the coffee machine”.</li><li>In the evening: an after party with pizzas, craft beers, cocktails and a DJ set by Tremolo.</li></ul><p><strong>Lineup :</strong></p><p>— 09:00 · Arthur Hilhorst (Lovable): Too fast to think<br> — 10:00 · Noé Chagué and Maxime Podgorski (Éno): The future of design with agentic AI<br> — 11:00 · Anthony Lefebvre and Mathéo Evrard: We broke the Double Diamond!<br> — 13:30 · Karine Dusart and Renaud Dorizon: A CPO’s first 60 days (panel)<br> — 14:40 · Anicet Bart (Solfy): Now everyone can build. But how do you win and keep users?<br> — 15:15 · Stéphane Cruchon: The Design Sprint in the age of AI: deciding fast, and right<br> — 16:15 · Roger Ingouacka (Petit Bambou): Lessons learned: the life of a real Product Builder!</p><p><strong>What I take away from each talk?</strong></p><h3>Arthur Hilhorst (Lovable): too fast to think</h3><p>Arthur is a product designer at Lovable, the Swedish start-up that builds applications with AI: about 60 million projects created, more than a million a week, and close to a billion visits a month on the generated apps. His talk came down to three words: <strong>speed, context, intent</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-uu4dWAtKgI95PHAgrHt_A.jpeg" /></figure><ul><li><strong>Speed.</strong> He asked the room to pick between thinking and speed: barely 20 to 30% of hands went up for thinking. Everyone feels the tension.</li><li><strong>Context.</strong> Running 10, 20 or 30 agents in parallel creates a huge cognitive load. At Lovable, a “second brain”, an internal LLM connected to the codebase and to user feedback, holds the memory for them.</li><li><strong>Intent.</strong> AI tools push us into the solution space too quickly, skipping the exploration of the problem.</li></ul><p>On the organisation side: designers ship straight into the codebase, 70 to 90% of pull requests are generated by AI, and there is no design system in Figma any more. The source of truth is the code.</p><p>The part that struck me most was about cognitive load. We used to juggle Jira, Miro and Figma. Now we juggle 10 tabs, 10 agents and 10 contexts to keep in mind. Arthur has tried it: you can run 20 or 30 agents, but not do deep thinking over several days at the same time. Agents generate so much, so fast, that your brain can no longer keep track and drops out. And losing control is exactly when you start skipping the details. Hence the “second brain”: an external memory that holds the context for you. With one limit, given in answer to a question from the room: too much context gives generic answers, so start from one source and balance it afterwards.</p><blockquote><em>“You only have one brain, so you can’t do everything.”</em></blockquote><blockquote><em>“If everyone starts at the same level, without intent, things really start to look alike.”</em></blockquote><blockquote><em>“If you wonder what craft and taste mean, look at Apple.”</em></blockquote><p>His example stuck with me: the flyers in the supermarket, all generated with ChatGPT, where the protest poster looks like the piano lessons ad. Craft and taste will make the difference, hence a role emerging at Lovable: artistic profiles who understand code and iterate thousands of times on a visual detail.</p><h3>Noé Chagué and Maxime Podgorski (Éno): an “AI ready” design system</h3><p>Éno helps companies automate their business (invoicing, website and more). Its design team works directly in the code. Their case study: the 17 November launch week, with <strong>143 screens generated in one week</strong>, about sixty of them in less than 24 hours.</p><p>The starting point: our first experiences with agents are disappointing. Components recreated, tokens ignored, spacing rules forgotten. For the Éno team, an agent asks itself three questions as soon as you give it an intent: what do I use? What am I allowed to do? And what do I do when the answer doesn’t exist? The third one is what matters.</p><p>Their answer:</p><p>The outcome of the week: 80% of the screens shipped after review, 20% revealed gaps in the system. And those 20% were the most valuable part.</p><blockquote><em>“What’s interesting isn’t the 140 screens shipped, it’s the gap, where the system fell short.”</em></blockquote><blockquote><em>“Once something is clear to an agent, it’s clear to a human too.”</em></blockquote><blockquote><em>“Good designers have years ahead of them.”</em></blockquote><p>Their maturity scale is simple. Level 1: the agent guesses which component to use. Level 2: documented metadata, tokens and intents, the agent follows your conventions. Level 3, “AI ready”: the system talks to you and tells you when it falls short.</p><h3>Anthony Lefebvre and Mathéo Evrard (Kiabi): we broke the Double Diamond</h3><p>The topic came up by chance, in a corridor: Kiabi’s second-hand clothing collection, which works very well in stores and in the app (four times the web volume), had to evolve so customers could send their clothes from home. The usual answer came: “we don’t have time”. They set themselves the challenge of answering it in less than 48 hours.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/0*l-QYbFQd6tSudXzp.jpeg" /></figure><p><em>The compressed Double Diamond: six phases, sixteen hours of work over 48 hours.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/0*XFKuqA7LSTqP1vz6.jpeg" /></figure><p><em>80% for AI (screens, questionnaire, synthesis), 20% for the team: the details that make the difference.</em></p><p>The most interesting part is what AI did not compress. Three times, the clock slowed down: <strong>choosing</strong> the direction, <strong>deciding</strong> between options, <strong>going to see</strong> people. A quarter of the time, and most of what they learned.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/0*4OrW2R-yCI5quqTp.jpeg" /></figure><blockquote><em>“We’ve all used that excuse, but today, with AI, ‘we don’t have time’ doesn’t necessarily hold any more.”</em></blockquote><blockquote><em>“AI only compresses what is already solid.”</em></blockquote><blockquote><em>“We didn’t save time to go faster, but to spend more time on what really matters.”</em></blockquote><h3>Karine Dusart and Renaud Dorizon: a CPO’s first 60 days</h3><p>A panel on a topic rarely covered at conferences: bringing a product culture into long-established mail-order companies, where “for a very long time, the product was the catalogue”. Renaud works at La Redoute, where product reports to marketing; Karine reports to an IT department.</p><ul><li><strong>The first 60 days.</strong> In theory: observe, listen, treat your team as a product. In practice: HR, budget and operational decisions to make from the first week. Three values set from day one: kindness, transparency, trust.</li><li><strong>Project or product?</strong> No opposition. Big projects reassure shareholders and defend budgets; the challenge is to bring discovery and vision into them.</li><li><strong>Marketing or IT?</strong> Marketing gives you the budget levers, IT gives you proximity with tech but makes it harder to engage the business. There is no universal right answer.</li><li><strong>Hiring.</strong> For a PM: soft skills, posture and command of their insights rather than hard skills alone. For a designer: a product vision beyond the mock-up, and a tech sensitivity that has become essential.</li><li><strong>AI.</strong> Security, licensing and budget constraints make it impossible to move as fast as the morning’s demos. At La Redoute, one PM is now a full-time AI builder, with ambassadors in every team.</li></ul><blockquote><em>“The business is going to be your BFF” (your best ally).</em></blockquote><blockquote><em>“Culture isn’t built in a month, maybe not even in a year. It’s what remains once you’ve left a company.”</em></blockquote><blockquote><em>“We don’t push mock-ups, we push experiences.”</em></blockquote><blockquote><em>“Accepting, as a leader, that my team sometimes loses time exploring AI. And I’m not afraid to say it.”</em></blockquote><p>Karine also quoted Marty Cagan’s <em>Transformed</em>: the most important thing is not adopting a process, but earning the business’s trust. And Renaud shared his prioritisation filter: “what is my added value on this topic?”. His direct reports are there so he can trust them.</p><h3>Anicet Bart (Solfy): everyone can build, now what?</h3><p>Anicet founded Solfy, a music theory learning app incubated on the campus. His message: building is no longer the problem, finding and keeping users is.</p><ul><li>Put the product in users’ hands from the first weeks.</li><li>Support them: a new product needs time to be adopted, and support is fundamental.</li><li>Show concrete uses, especially on Instagram.</li><li>Own the marketing: more than a third of the project budget goes to promotion.</li><li>Go and meet people, at trade shows, on the phone, all the time.</li><li>Structure the data the product generates from the start: it’s an asset AI can use tomorrow.</li></ul><p>On the business model: a €40 yearly subscription per family, which local councils can pay for.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/0*E08itoGqQDCVcPZU.jpeg" /></figure><blockquote><em>“You don’t retain a user. You make them succeed.”</em></blockquote><h3>Stéphane Cruchon: the Design Sprint in the age of AI</h3><p>Stéphane has run Design Sprints since 2015, alongside Jake Knapp, who invented the method at Google Ventures. More than 100 companies coached in ten years, from start-ups to large groups.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/0*iB4X3LS5K916j1CW.jpeg" /></figure><p>His question: why still block five days when AI generates a prototype in a few minutes? Because a new project drowned in calendars dies, and because decisions are made by humans, not by agents.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/0*FfoaEPYf20QEdnrI.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/0*AO_niRyng1U8xXL9.jpeg" /></figure><blockquote><em>“It takes less than a minute to make something, but it takes at least five days to make a real business decision.”</em></blockquote><blockquote><em>“Days 1 to 3 are context engineering. Done by humans.”</em></blockquote><blockquote><em>“Design isn’t dead: you need it to decide, and to believe in the vision.”</em></blockquote><p>His examples: an agent that analyses customer service calls and fills in the CRM, prototyped in a day and launched in less than six months; a heritage brand renamed two weeks after a sprint, because the tests showed the name was a problem; a Swiss neobank that grew to 800,000 customers. Each time, the emotional value of the process matters: people defend what they built together.</p><h3>Roger Ingouacka (Petit Bambou): the life of a Product Builder</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/0*FcifQUvBBmeZ5RYp.jpeg" /></figure><p>Roger, a developer turned Product Builder, has spent two years on Petit Bambou’s new vertical: <strong>Mouvement</strong>, released in September. The observation: people in their fifties in pain everywhere, with no condition, just after 15 to 20 years of sitting down. The goal: get your body moving every day, with no pressure to perform or to look a certain way.</p><p>All this with a team of about twenty people, without dedicated squads, and a very fair question from the audience: how do you stop an AI from chaining dangerous movements?</p><h3>The threads running through the day</h3><p><strong>1. Humans keep what can’t be compressed.</strong> Kiabi measured it, Stéphane Cruchon built it into his method: choosing, deciding and going to see people don’t compress. AI takes the rest.</p><p><strong>2. The source of truth becomes the code, and context becomes a product.</strong> Lovable and Éno no longer have a design system in Figma. Their design system talks to agents, and context is built like a tool: second brain, MCP, CI, changelog.</p><p><strong>3. Speed without intent produces products that look alike.</strong> Craft, taste and decision-making become the difference again.</p><p><strong>4. Building is no longer enough.</strong> Anicet Bart said it bluntly: now that everyone can build, the challenge is to win users and make them succeed.</p><h3>How I felt about it</h3><p><strong>Very few tech people in the room.</strong> And yet, at a time when AI augments our skills and widens what we can do, these topics concern us more and more: agent-driven design systems, CI as a safety net, builders shipping into the codebase.</p><p><strong>Great organisation, and a conference designed as an experience.</strong> Lots of humour, humanity, lightness and simplicity. A perfect example: not-so-comfortable chairs, introduced with humour, and cushions printed with the Nord Star Product logo for the occasion. My favourite part: the welcome tote bag held a bottle of tripel beer with the Nord Star Product logo.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ND3GeIk6W2zHE5o-.jpeg" /></figure><p><em>Lunch, by Barkette.</em></p><p>Happy to see Rémi again, whom I hadn’t seen in ages, behind his food truck, Barkette. Thanks to Xavier Koma and Teddy Lecaillon, whom I was glad to see again, and to the whole organising team.</p><p><strong>Very good speakers and talks</strong>, and real conversations with some of them, especially Anicet Bart and Roger Ingouacka.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/0*3XjUoNUJqOIbt8n1.jpeg" /></figure><p><em>The after party, with Tremolo on the decks.</em></p><p><strong>On video</strong>, to get a feel for the atmosphere: <a href="https://lnkd.in/p/eFRqN2wP">Stéphane Cruchon’s recap of the event</a>, and <a href="https://lnkd.in/p/eV7VPs9S">another nice video of the day, by Romain Dao from The Cacatoès Theory</a>.</p><h3>In short</h3><p>Nord Star Product Aurora showed teams that really use AI, in very different contexts: a Swedish start-up, a team that merges up to 150 pull requests a day, a Kiabi product team, two CPOs from heritage companies, a founder, a Product Builder. Their shared message: AI speeds up everything that is already solid, and leaves us the choices, the decisions and the fieldwork.</p><p>To go further:</p><ul><li>The <a href="https://nordstarproduct.com/lineup">full Aurora edition programme</a>.</li><li><a href="https://lnkd.in/p/eFRqN2wP">Stéphane Cruchon’s video recap</a> of the event, and <a href="https://lnkd.in/p/eV7VPs9S">another video of the day, by Romain Dao</a>.</li><li>Jake Knapp’s book <em>Sprint</em>, and Marty Cagan’s <em>Transformed</em>, quoted during the panel.</li></ul><p>Were you there? Which talk stayed with you the most?</p><h4>Sources</h4><ul><li><a href="https://nordstarproduct.com/lineup">Nord Star Product, Aurora edition programme</a></li><li>My notes, the automatic transcripts of the talks and my photos of the day</li></ul><p><em>Originally published at </em><a href="https://maxime-lenne.fr/en/blog/nord-star-product-aurora-2026/"><em>https://maxime-lenne.fr</em></a><em> on October 24, 2026.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=53bbe737252d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[In the Loop Berlin 2026: key takeaways from the n8n opening keynote]]></title>
            <link>https://medium.com/@MaximeLenne/in-the-loop-berlin-2026-key-takeaways-from-the-n8n-opening-keynote-34b22014aa5a?source=rss-7ea898cb1bf4------2</link>
            <guid isPermaLink="false">https://medium.com/p/34b22014aa5a</guid>
            <category><![CDATA[n8n]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[no-code]]></category>
            <category><![CDATA[automation]]></category>
            <category><![CDATA[n8n-workflow]]></category>
            <dc:creator><![CDATA[Maxime Lenne]]></dc:creator>
            <pubDate>Tue, 20 Oct 2026 13:01:04 GMT</pubDate>
            <atom:updated>2026-10-27T22:32:12.837Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8JfK61rFfAJUiPJprhcw3w.jpeg" /></figure><p><strong>Building a workflow with AI now takes 40 seconds. Checking it takes 40 minutes.</strong> That is the line I keep from the opening keynote of In the Loop, the first n8n conference, on 14 October 2026 in Berlin.</p><p>A week ago, I wrote <a href="https://lnkd.in/p/eVAN7jXS">on LinkedIn</a> what I was watching for: MCP, more autonomous yet controllable agents, human-in-the-loop. I shared <a href="https://lnkd.in/p/eSKUjft2">my first impressions</a> on the day. Here is the long version: the figures, Jan Oberhauser’s vision, David Roberts’s product announcements, and what it all means for people who build with n8n.</p><h3>In the Loop: n8n comes home</h3><p><a href="https://n8n.io/intheloop/">In the Loop 2026</a> is the first major n8n conference. It took place on 13 and 14 October 2026 at Wilhelm Studios in Berlin, the city where n8n started in 2019. The event website sums it up in one line: “We’re coming home”.</p><ul><li><strong>13 October, Partner Day</strong>: invite-only day for partners and ambassadors.</li><li><strong>14 October, Ecosystem Day</strong>: open to everyone, with keynotes, talks, workshops, an expo and an open mic.</li><li><strong>Capacity</strong>: 1,200 attendees at most, plus the live stream.</li><li><strong>Sponsors</strong>: SAP, Accenture, Capgemini, Avanai, Exxeta and SVA.</li></ul><p>The opening keynote on 14 October had two parts. Jan Oberhauser, founder and CEO, covered the vision and the “why”. David Roberts, VP Product &amp; Design, then walked through the product: what has shipped, what is coming and what is still being explored.</p><h3>The key announcements</h3><h3>n8n in numbers</h3><p>Jan Oberhauser opened with the scale reached since the first line of code, written in Berlin in 2019.</p><ul><li><strong>6.7 million</strong>: active workflows per month</li><li><strong>60%</strong>: Fortune 500 companies using n8n</li><li><strong>€5.2 billion</strong>: valuation after the SAP investment</li><li><strong>519,000</strong>: community members</li><li><strong>13,000</strong>: published templates</li><li><strong>2,300</strong>: integrations</li><li><strong>340</strong>: ambassadors</li><li><strong>207,000</strong>: GitHub stars (top 30 of all time)</li><li><strong>400</strong>: active partners</li></ul><h3>The vision: “More capable AI should create more capable people”</h3><p>Jan asks a single question: as AI gets smarter every month, what happens to people? n8n bets on the path where humans understand more and stay in control.</p><p>His argument is a chain: <strong>visibility → understanding → control → agency</strong>. Remove the first link and everything falls. Hence four questions before trusting an automation: what goes in, what happens, what comes out, and who is accountable?</p><p>Two ideas follow:</p><ul><li><strong>The UI is the trust layer.</strong> It does not fade away with AI; it makes actions, boundaries and responsibility readable.</li><li><strong>The more critical the process, the more you need to see it.</strong> A personal task: the result is enough. A shared team process: the key steps. A business-critical operation (payroll, customer onboarding, anything regulated): every step.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wDJKhxCSRtTkrO7kJZyocQ.jpeg" /></figure><p><em>Visibility follows criticality: the result, the key steps, or every step.</em></p><p>The new positioning: “AI makes intelligence possible. n8n makes intelligence operational.” n8n wants to grow from a product into an <strong>operating system for AI-powered work</strong>. Critical work needs four building blocks: agents (judgment and language), deterministic software (one right answer), automation (repeatable steps at any volume) and people and teams (ownership and the final call).</p><h3>A rebrand and n8n Labs</h3><p>Max, “the original Flowgrammer”, unveiled the new identity. The old logo celebrated the nodes; the new one celebrates the space between them, where the connection happens. The colour stays.</p><p>Jan also introduced <strong>n8n Labs</strong>, a small team exploring new products under three rules: speed and signal over polish, build for the capability curve rather than the current model, and default to shutting things down. A first Labs product was teased at the end of the keynote, with a waiting list.</p><h3>What will never change</h3><ul><li><strong>Community Edition</strong>: free, self-hosted, with the source open to read.</li><li><strong>Model choice</strong>: your models, your providers.</li><li><strong>Equal treatment</strong> of free and paying users.</li><li><strong>50% of engineers</strong> stay focused on the n8n core.</li></ul><h3>The product, by David Roberts</h3><p>David Roberts structured his part around the lifecycle of an automation: <strong>build, audit, deploy, monitor, govern</strong>.</p><p><strong>Already shipped:</strong></p><ul><li><strong>Agents (October)</strong>: Agents taken out of workflows and put on the same level, better MCP and Slack support. <em>Result: 20% more likely to reach production.</em></li><li><strong>AI Assistant</strong>: Builds workflows from natural language, on Cloud and self-hosted. <em>Result: 1.6M conversations, 800,000 workflows, 8 in 10 users satisfied.</em></li><li><strong>MCP for building</strong>: Connect Claude, Codex or another coding agent to n8n. <em>Result: 3x more workflows, 3.4x more likely to succeed.</em></li><li><strong>n8n Connect</strong>: Sets up selected credentials automatically. <em>Result: used by 25% of users, 25% higher success rate.</em></li></ul><p><strong>Coming next:</strong></p><ul><li><strong>Simpler evaluations.</strong> Only 15% of teams run evals, yet teams with rigorous evals have a 70% chance of excellent reliability (32% without). n8n wants to surface only the business decisions and handle the rest with AI.</li><li><strong>A new execution engine</strong>, rewritten from the ground up: progress saved after every step, a reconciliation suite for stuck jobs, automatic retries only on steps marked safe, sandboxed code, worker pools, resource isolation.</li><li><strong>Manifest</strong>, a separate observability tool for multi-instance setups: what exists, what runs, who owns it, which systems are touched. In testing, open to multi-instance users.</li><li><strong>Governance</strong> along four axes: access, guardrails, ownership and quality, security. Eventually with agent identities of their own, narrow, attributable and revocable, instead of borrowing a person’s login.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*a_UAtbPeLKKWoJLDy3xloA.jpeg" /></figure><p><em>First look at Manifest, n8n’s multi-instance observability.</em></p><h3>What it means for n8n builders</h3><p><strong>1. Agent first, workflow second.</strong> The agent becomes the entry point: you describe an outcome, it finds the path. But David Roberts is clear: an agent doing the same thing 10,000 times a day is just a very expensive workflow. The rule of thumb: if the thinking happens once, at design time, it is a workflow; if it happens on every run, it is an agent.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nbahUDGgGf9Ry7CjNTpQaQ.jpeg" /></figure><p><em>Agent-first, workflow second: agents as the entry point, deterministic logic wherever the work requires it.</em></p><p><strong>2. Make deterministic whatever can be.</strong> Reliability multiplies along a chain: 10 steps at 95% each give a system that is 60% reliable (0.95 to the power of 10). The advice: take business logic out of the agent and wrap it in deterministic tools. The example given: a single “onboard a user” tool rather than three Slack, Google Workspace and Salesforce tools left to the agent’s choice.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GJVWJZ33gWCXPkPVOJjQIg.jpeg" /></figure><p><em>agent “figures it out”; the workflow “does exactly what you said, every time”.</em></p><p><strong>3. The bottleneck has moved from building to auditing.</strong> You build in 40 seconds and check for 40 minutes. There are two ways to audit: inspection (reading the logic, the job of the canvas) and testing (running it and checking the results, the job of evals). Of 1,127 users surveyed, 55% said n8n would lose much of its meaning without the canvas. The canvas stays.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b-rr0LYtFdX_xWcejIXrtQ.jpeg" /></figure><p><em>The new bottleneck: the agent builds in 40 seconds, you check for 40 minutes.</em></p><p><strong>4. Trust is earned in stages.</strong> First the agent suggests and the human acts. Then the agent acts and the human approves spending. Finally the human sets a budget and walks away. Budgets are a partial answer to “approval fatigue”: clicking “OK” a hundred times without reading.</p><p><strong>5. Governance is no longer for large companies only.</strong> Soon there will be more agents than employees, even in small businesses. Shadow IT is no longer a spreadsheet macro; it is a cluster of agents with access to sensitive systems.</p><p>Two customer stories show the range. Mercedes-Benz chose n8n as its global automation platform, layered and cloud agnostic. At the other end, a small US agency turns WhatsApp voice notes into CRM updates: from 4–5 minutes down to 10–20 seconds per note, and a sales cycle cut from 62 to 44 days.</p><h3>My take: the real battle is visibility in production</h3><p>What stands out to me is that n8n did not give a “magic AI” keynote. The common thread, from Jan’s talk to David’s roadmap, is trust: seeing every step, auditing, monitoring, governing. Building with AI has become easy; knowing what runs on a Sunday night, when nobody is watching, has not.</p><p>That is exactly the problem I work on with <a href="https://n8n-ninja.app/en/?utm_source=medium&amp;utm_medium=referral&amp;utm_campaign=blog-intheloop">N8Ninja</a>, my app to manage and monitor n8n automations from Apple devices. I released <a href="https://lnkd.in/p/eXfQYWw9">version 1.2.1</a> for the occasion. It answers several of the questions raised on stage:</p><ul><li><strong>On stage</strong>: “The more critical, the more you need to see each step”. <strong>In N8Ninja</strong>: Debug logs with the JSON input and output of every node.</li><li><strong>On stage</strong>: Monitoring: keeping an eye on what runs. <strong>In N8Ninja</strong>: Status, duration and errors of each execution; workflow health over 48 h and 7 days.</li><li><strong>On stage</strong>: Trusting it runs when nobody is watching. <strong>In N8Ninja</strong>: Failure alerts, with background notifications on Mac.</li><li><strong>On stage</strong>: Security, the fourth pillar of governance. <strong>In N8Ninja</strong>: Security audit with a risk estimate.</li></ul><p>N8Ninja runs on iPhone, iPad, Mac, Apple Watch and Apple TV, with n8n Cloud as well as self-hosted instances. Manifest targets observability for large multi-instance organisations; N8Ninja targets the builder who wants the state of their workflows in their pocket.</p><h3>In short</h3><p>n8n set a clear course in Berlin: AI builds, people understand and decide. Agents as the entry point, workflows for reliability, and a big push on auditing, the execution engine, observability and governance. The Community Edition and model choice remain guaranteed.</p><p>To go further:</p><ul><li>Watch <a href="https://www.youtube.com/watch?v=KQRe5UurLQ0">the full keynote on YouTube</a>.</li><li>Try the AI Assistant or connect your coding agent through MCP, if you have not yet.</li><li>Get in touch with n8n to test Manifest if you run several instances.</li><li>Monitor your workflows from your iPhone, Mac or Apple Watch with <a href="https://apps.apple.com/us/app/id6746260475?pt=127799253&amp;ct=medium-intheloop&amp;mt=8">N8Ninja</a>.</li></ul><p>Which announcement matters most to you? Tell me in the responses.</p><h3>Sources</h3><ul><li><a href="https://n8n.io/intheloop/">In the Loop 2026, official website</a></li><li><a href="https://www.youtube.com/watch?v=KQRe5UurLQ0">Livestream: Opening keynote and Product direction (YouTube)</a></li><li><a href="https://www.linkedin.com/posts/n8n_intheloop2026-activity-7516128721997787136-7HUZ">n8n on LinkedIn: the day in pictures</a></li><li><a href="https://www.linkedin.com/posts/n8n_intheloop2026-activity-7516165060868591618-KGJJ">n8n on LinkedIn: community moments</a></li><li>My notes, the keynote transcript and my captures of the live stream</li></ul><p><em>Originally published on the </em><a href="https://n8n-ninja.app/en/blog/n8n-in-the-loop-berlin-2026-keynote/?utm_source=medium&amp;utm_medium=referral&amp;utm_campaign=blog-intheloop"><em>N8Ninja blog</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=34b22014aa5a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Bumblebee bot : générer des véhicules robots]]></title>
            <link>https://medium.com/mobilityz/bumblebee-bot-generate-random-journey-f86e9ad35436?source=rss-7ea898cb1bf4------2</link>
            <guid isPermaLink="false">https://medium.com/p/f86e9ad35436</guid>
            <category><![CDATA[transport]]></category>
            <category><![CDATA[transportation]]></category>
            <category><![CDATA[google-maps]]></category>
            <category><![CDATA[mapbox]]></category>
            <category><![CDATA[nodejs]]></category>
            <dc:creator><![CDATA[Maxime Lenne]]></dc:creator>
            <pubDate>Wed, 30 Nov 2016 11:09:58 GMT</pubDate>
            <atom:updated>2016-11-30T11:09:58.842Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7FVvDs0L-j2J77i1MvLiLA.jpeg" /></figure><p>Depuis la naissance d’<a href="https://ecota.co/">EcoTa.co</a> nous avons une problématique lors de nos test en local ou sur l’environnement de staging : avoir <strong>des remontés de données fictives</strong> pour simuler le matériel que possède notre flotte de chauffeur <em>VTC</em>.</p><p>En effet les chauffeurs d’<a href="https://ecota.co/">EcoTa.co</a> peuvent travailler avec différent types de matériels qui permettent de les connecter au système : des GPS tomtom professionnels et des applications natives iOS et android. Ces matériels nous <strong>envoie en permanence leurs données:</strong> la géolocalisation, l’orientation, la vitesse, le mode de déplacement, la disponibilité et permettent également de remonter les différents statuts associés à une mission que nous leurs envoyons (dans notre cas un transport de personne). Bien sûr nous disposons de matériels au bureau, mais ceux-ci ne sont malheureusement pas en situation réelle, à savoir en mouvement.</p><p>La team <a href="http://mobilityz.io/">mobilityz</a> a donc créé le projet <a href="https://bumblebee-bot.herokuapp.com/">Bumblebee bot</a>, dont le but est de <strong>générer des matériels fictifs simulant une flotte de chauffeur</strong> <em>VTC</em>. Pourquoi ce nom ? Il se comprend facilement compte tenu de notre objectif : générer des robots qui simulent le déplacement de véhicules. Quoi de mieux que les transformers pour représenter le projet :)</p><p>Nous avons profité de notre participation au salon de la <a href="http://jdmc.info/">Journée De la Mobilité Connectée</a> pour présenter ce projet. La <a href="https://twitter.com/search?q=%23JDMC16&amp;src=typd">#JDMC16</a> c’est un événement dédié à la mobilité de demain organisé par <a href="https://medium.com/u/2af44a51f142">tymate</a> qui s’est déroulé à <a href="http://euratechnologies.com/">Euratechnologies</a> le 16 novembre dernier.</p><h3>De quoi notre robot est-il capable ?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*O1scWs246Zczz-bcV8OcoA.jpeg" /></figure><p>Comme notre ami bumblebee pas besoin de lui ajouter un module vocal, c’est pas vraiment ce qu’on lui demande. Par contre un certain nombre de fonctionnalités nous sont venues à l’esprit lors de ce projet, nous en avons traitées certaines mais d’autres seront pour un prochain sprint.</p><p>L’idée de départ est de pouvoir créer un bot qui possèdent <strong>n chauffeurs dans un polygone</strong> dessiné par l’utilisateur. Ce robot va <strong>générer des déplacements aléatoires</strong> pour chaque chauffeurs dans cette zone, pour ensuite envoyer régulièrement sur une API les données fictives décrites plus haut.</p><p>Voici les principales idées / fonctionnalités auxquels nous avons pensées pour ce projet :</p><ul><li>Générer des points aléatoires dans un polygon</li><li>Générer un ensemble de points pour un trajet</li><li>Limiter les trajets à l’intérieur du polygone</li><li>Générer l’orientation sur chaque point du trajet</li><li>Gérer la vitesse et la précision entre chaque points</li><li>Gérer les limitations de vitesses</li><li>Se servir du trafic en temps réel</li><li>Pouvoir accepter des missions et dérouler les statuts</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fw9I2mbXBq-nBk_XqJ6oHQ.jpeg" /><figcaption>Allez sortons l’artillerie ;)</figcaption></figure><h4>NodeJS</h4><p>Nous avons créé le projet en nodeJS, cela nous a permis de tester les nouveautés principales d’ES6 (Javascript version 2015) : les promises.<br>Si vous avez déjà fait du Javascript vous devez connaître les callbacks, on passe en paramètre d’une fonction un callback qui sera appelé à la fin de l’exécution de la fonction.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/f1580eb59405f5d9935c6bdd6f4523d6/href">https://medium.com/media/f1580eb59405f5d9935c6bdd6f4523d6/href</a></iframe><p>Ici on attends que la base de données renvoie les données du bot avant de l’afficher. On remarque rapidement qu’avec ce genre de techniques on peut vite tomber dans ce qu’on appelle le “<em>callback hell</em>”, un callback dans un callback dans un callback … On ne sait plus dans quel ordre les fonctions sont exécutées, ni ce qu’elles font.</p><p>Une partie de la solution vient des promesses, elles permettent entre autre d’améliorer grandement la lisibilité du code. Une promesse c’est un objet JS qui pourra un jour potentiellement avoir une valeur.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/17165fa7a5936d1381a1fc1132cffa2c/href">https://medium.com/media/17165fa7a5936d1381a1fc1132cffa2c/href</a></iframe><p>Et on peut les chaîner, par exemple :</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/702232ff20189000ebd78e66c1cd8667/href">https://medium.com/media/702232ff20189000ebd78e66c1cd8667/href</a></iframe><h4>Cartographie et calcul d’itinéraire</h4><p>Pour <strong>générer un point aléatoire dans un polygone</strong>, nous recréons un point aléatoire dans le rectangle contenant le polygon (bounding box) tant que celui-ci n’est pas à l’intérieur de notre polygone.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/3978c47df1a6ff48be26b0c58700df6c/href">https://medium.com/media/3978c47df1a6ff48be26b0c58700df6c/href</a></iframe><p>L’<a href="https://developers.google.com/maps/documentation/directions/?hl=fr">API Directions de Google Maps</a>, nous permet de récupérer une ligne polygonale contenant les points nécessaires pour tracer la route entre les points que nous avons générés aléatoirement. Malheureusement l’API ne nous retourne pas un nombre suffisant de points pour simuler des déplacements avec précisions. Nous avons du coup regardé la nouvelle <a href="https://developers.google.com/maps/documentation/roads/intro?hl=fr">API Roads de Google</a>, plus précisément la méthode <em>snapToRoads</em> mais qui ne correspondait pas à notre besoin. Nous avons finalement implémenté notre propre code afin d’<strong>affiner le nombre de points</strong> en fonction de la distance entre ceux-ci et la précision que nous souhaitons.</p><p>Pour générer l’<strong>orientation entre deux points</strong>, nous avons trouvé <a href="http://www.movable-type.co.uk/scripts/latlong.html">cet article</a> qui nous fournit une formule pour ce calcul.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/b5a64f3f364b57892acbf0cabd8ef326/href">https://medium.com/media/b5a64f3f364b57892acbf0cabd8ef326/href</a></iframe><p>Pour la partie front et l’animation de la carte nous avons utilisé <a href="https://www.mapbox.com/">Mapbox</a> (que nous apprécions particulièrement chez <a href="http://mobilityz.io/">mobilityz</a>) avec la librairie Leaflet. Cette librairie permet, entre autres, de dessiner des polygones, d’afficher des trajets et d’animer des marqueurs.</p><h4>Jobs</h4><p>Nous avons ajouté au projet un système de jobs avec <a href="http://automattic.github.io/kue/">Kue</a> et Redis pour pouvoir réaliser les opérations de générations de points, de calculs d’itinéraires, générations d’orientations et remontées des données fictives sur notre API et websocket.</p><p>Cela nous permet pour le calcul d’itinéraire, qui peut avoir un temps d’exécution long, d’être sûr de ne pas bloquer l’exécution du processus principal et également facilement répartir la charge.</p><p>Pour la remontée de données on utilise la possibilité de mettre un délais à chaque job, ainsi on peut directement créer toutes les remontées d&#39;un trajet et leur mettre un délais croissant. Ainsi le worker va exécuté automatique les jobs avec le bon décalage.</p><h4>Websocket</h4><p>Pour la démo des <a href="https://twitter.com/search?q=%23JDMC16&amp;src=typd">#JDMC16</a> nous avons mis en place un <em>websocket</em> avec <a href="http://socket.io/">socket.io</a> afin de récupérer facilement côté front les nouvelles remontés de nos véhicules fictifs et pouvoir les afficher directement dans notre projet.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*HLcSxeFsD0CLH8t7MVT8Jw.gif" /><figcaption>Transformation de Bumlebee en camaro</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/854/1*CL7AL4qL5Dy4cubce0qxeg.png" /><figcaption>Dix chauffeurs sur Lille</figcaption></figure><p>Merci à <a href="https://medium.com/u/d1c038d5932e">Gabriel de Vareilles</a> et <a href="https://medium.com/u/33c07e7ddd48">Jean Bertrand</a> qui ont co-écrit cet article et bien évidemment participé à la conception du projet.</p><p>La demo ➡️ <a href="https://bumblebee-bot.herokuapp.com/#">https://bumblebee-bot.herokuapp.com</a></p><p><strong>Si ce projet vous intéresse et si vous ayez des suggestions d’améliorations ou ajout de fonctionnalités n’hésitez pas à échanger avec nous ! </strong><a href="https://bumblebee-bot.herokuapp.com/"><strong>Bumblebee bot</strong></a><strong> peut être idéal pour tester votre système de gestion de flotte par exemple.</strong></p><p><a href="/service/mailto:hello@mobilityz.io">hello@mobilityz.io</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f86e9ad35436" width="1" height="1" alt=""><hr><p><a href="https://medium.com/mobilityz/bumblebee-bot-generate-random-journey-f86e9ad35436">Bumblebee bot : générer des véhicules robots</a> was originally published in <a href="https://medium.com/mobilityz">mobilityz</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Mobilityz : À la croisée des chemins entre deux expertises, le transport et la technologie]]></title>
            <link>https://medium.com/mobilityz/mobilityz-la-croisee-des-chemins-entre-deux-expertises-le-transport-et-la-technologie-db643eee7f30?source=rss-7ea898cb1bf4------2</link>
            <guid isPermaLink="false">https://medium.com/p/db643eee7f30</guid>
            <category><![CDATA[innovation]]></category>
            <category><![CDATA[transportation]]></category>
            <category><![CDATA[mobility]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[web-development]]></category>
            <dc:creator><![CDATA[Maxime Lenne]]></dc:creator>
            <pubDate>Tue, 04 Oct 2016 15:51:53 GMT</pubDate>
            <atom:updated>2016-11-04T13:25:22.267Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3sIUVzqEF3JxvhUCe1svtw.png" /></figure><h3><a href="http://mobilityz.io/">Mobilityz</a> : quand le transport et la technologie s’associent…</h3><p>Fin 2012 nous avons créé la société FactorApp SAS dont l’objectif premier était de développer <a href="https://ecota.co/">EcoTa.co</a>, une plateforme de mise en relation entre chauffeurs privé VTC et des utilisateurs particuliers ou professionnels. Depuis la création de l’entreprise et malgré les nombreuses sollicitations que nous avons reçues, nous avons bien souvent refusé d’effectuer des prestations informatiques, dans le but principal de poursuivre notre objectif de départ: développer EcoTa.co et à travers cela démocratiser le transport à la demande pour le rendre accessible à tous.</p><h3>Un sacré chemin parcouru</h3><p>Aujourd’hui plébiscité par des milliers d’utilisateurs et des centaines de chauffeurs dans toute la France, le service <a href="http://app.ecota.co/">EcoTa.co</a> a atteint une certaine maturité. En effet, la récurrence de nos utilisateurs et le nombre de clients ou d’entreprises utilisant le service sont la preuve d’un certain succès de cette aventure entrepreneuriale menée par deux “newbies” de l’industrie du transport.</p><p>Comme toutes les plateformes d’intermédiation, nous avons du faire face à des développements technologiques très couteux depuis le départ pour automatiser (faire mieux, avec moins, et plus rapidement) et assurer une excellence opérationnelle du service. Aujourd’hui notre système d’information, qui continue néanmoins d’évoluer chaque jour, est “mature” et surtout scalable, c’est-à-dire qu’il peut faire face à un volume beaucoup plus important.</p><p>Ce constat nous permet enfin de sortir la tête de l’eau et de nous tourner vers l’avenir. Forts de notre expertise dans le domaine de la mobilité et du transport et conscients des ressources techniques que nous avons aujourd’hui internalisées, il nous est apparu comme évident de lancer une nouvelle activité: Mobilityz.</p><h3>Mobilityz : ambition, vision &amp; ADN</h3><p>L’idée derrière <a href="http://mobilityz.io/">Mobilityz</a> est très simple: créer <strong>une agence spécialisée en innovation digitale dans le domaine de la mobilité et du transport des biens et des personnes</strong>, avec une équipe autonome.</p><p><a href="http://mobilityz.io/">Mobilityz</a></p><p>L’ADN actuelle de l’entreprise est : <strong>l’innovation, la technologies, la mobilité et le transport</strong>. Celle-ci ne changeras pas, bien au contraire, nous aurons enfin l’opportunité de la développer et les prestations sur lesquelles nous travaillerons ‘colleront’ à cet ADN.</p><p>Nos objectifs en créant <a href="http://mobilityz.io/">Mobilityz</a> :</p><ul><li>Toujours rester au plus proche de l’innovation</li><li>Recruter et agrandir notre équipe technique</li><li>Partager nos compétences et enrichir notre expertise</li></ul><h3><a href="http://mobilityz.io/">Mobilityz</a> une agence déjà sur les rails !</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*WqpD7cZPM4UqwxpW.jpg" /><figcaption>toute nouvelle génération de trains à grande vitesse d’Alstom pour Amtrak</figcaption></figure><p>Des acteurs du transport nous font déjà confiance : SNCF, Tomtom, Brake, Bump, i-viatic, i-trans… Des expérimentations existantes <a href="http://lesretards.com/">lesretards.com</a> (Une API et une application web permettant de détecter les retards de train et d’avion en temps réel), <a href="https://github.com/Ecotaco/tomtom_webfleet_connect_gem">une librairie ruby open source</a> pour une utilisation simplifiée de l’API Tomtom… Et une solide expérience dans le monde de la prestation de service.</p><p><strong>Un produit SAAS de gestion de flotte en cours de développement : </strong>il permettra la remontée de géolocalisation et de la disponibilité en temps réel de collaborateurs, gestion des ordres de missions, propositions de missions en temps réel aux collaborateurs les plus proches, et une API permettant de s’adapter à n’importe quel besoin…</p><p><strong>Partenaire développeur Tomtom .connect : </strong>Grâce au projet EcoTa.co nous avons acquis une solide expérience sur le matériel et les API de Tomtom Telematics, et sommes officiellement partenaire .connect Tomtom. Nous pouvons donc proposer des prestations de développement sur cette technologie en vue d’optimiser le système de gestion de flotte “webfleet” ou faciliter l’interfaçage avec un outil-métier existant.</p><p>Les premiers projets sont déjà à l’étude et cela nous conforte dans notre idée qu’il y a un vrai besoin dans cette industrie.</p><p><strong>Que vous soyez transporteur, ambulancier, collectivité, startup, gestionnaire de flotte… </strong><a href="/service/mailto:hello@mobilityz.io"><strong>Discutons</strong></a><strong> dès maintenant des solutions innovantes que nous pouvons vous apporter !</strong></p><p><a href="/service/mailto:hello@mobilityz.io">hello@mobilityz.io</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=db643eee7f30" width="1" height="1" alt=""><hr><p><a href="https://medium.com/mobilityz/mobilityz-la-croisee-des-chemins-entre-deux-expertises-le-transport-et-la-technologie-db643eee7f30">Mobilityz : À la croisée des chemins entre deux expertises, le transport et la technologie</a> was originally published in <a href="https://medium.com/mobilityz">mobilityz</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>