{"head":{"title":"Building Pylon - AI-Native Insurance Operations","description":"Replaced fragmented policy spreadsheets and disconnected claims tools with a unified AI-driven workspace, designed and shipped in 8 weeks via clickable…","canonical":"https://oceanlab.in/case-studies/pylon"},"study":{"slug":"pylon","industry":"insurance","projectType":"Insurance operations console","featured":true,"country":"us","tags":["insurtech","ai","saas","enterprise"],"compliances":["SOC 2 Type II","HIPAA-aligned","GDPR","ISO 27001"],"cover":{"src":"/pylon-1.png","width":1440,"height":1207},"gallery":[{"src":"/pylon-1.png","width":1440,"height":1207},{"src":"/pylon-2.png","width":1440,"height":1207}]},"publishedDate":"2026-09-25","contentByLang":{"en":{"client":"Pylon","clientLocation":"USA","category":"InsurTech","duration":"Phase 1 (8 Weeks)","title":"Building Pylon - AI-Native Insurance Operations from Prototype to Production","tagline":"AI-Native Insurance Operations Platform","description":"Replaced fragmented policy spreadsheets and disconnected claims tools with a unified AI-driven workspace, designed and shipped in 8 weeks via clickable AI prototype.","scope":"AI Prototyping, UI/UX Design, Development and Delivery","industry":"InsurTech","timeline":"8 Weeks","impacts":[{"value":"8 Weeks","label":"Prototype to production","description":"1 week scope, 2 weeks clickable AI prototype, 5 weeks production build."},{"value":"147K+","label":"Policies in one console","description":"Active portfolio rendered in a single workspace with real-time risk scoring."},{"value":"AI-Native","label":"Built into the workflow","description":"Per-policy AI insights, anomaly detection, and a chatbot trained on policy context."},{"value":"Single team","label":"Zero handoff","description":"AI prototyping, UI/UX design, and engineering shipped under one accountable team."}],"problemShort":"Pylon's team was managing thousands of policies across spreadsheets, separate claims tools, and inboxes - risk signals lived in three different systems and renewals were tracked manually.","solutionShort":"We delivered a unified AI-native workspace - one console for portfolio overview, policy detail, claims, and renewals - validated through a clickable AI prototype before any production code.","challengesHeading":"The Challenge: Fragmentation at Scale","challengesIntro":"An insurance ops team running thousands of policies needs a single source of truth. Pylon's didn't have one.","challenges":[{"title":"Fragmented Workflow","description":"Policies, claims, renewals, and risk lived in separate tools with no shared spine. Operators reconciled state across systems by hand."},{"title":"No Real-Time Visibility","description":"Risk scores and renewal urgency were calculated in batch jobs, leaving managers reacting to stale dashboards instead of acting on live signals."},{"title":"Slow Policy Onboarding","description":"Adding a new client policy meant updating four tools and a checklist - hours per policy and an audit trail that didn't quite line up."},{"title":"AI Without Context","description":"Existing analytics were static reports. There was no way for an underwriter to ask 'what changed on Henderson Corp last week?' and get an answer."}],"approachHeading":"Our Approach: 3 Phases in 8 Weeks","approach":[{"number":"01","title":"Discovery & Scope","description":"Week 1: embedded with underwriters and ops to map every workflow, click, and edge case. Defined the production target and success metrics on day five."},{"number":"02","title":"AI-Driven Clickable Prototype","description":"Weeks 2–3: built an interactive prototype using AI prototyping tools so the Pylon team could click through the actual product - including AI insights and chatbot - and refine it in real time before any production code."},{"number":"03","title":"Production Build & Launch","description":"Weeks 4–8: shipped the production console with portfolio overview, per-policy detail, claims hub, renewals tracker, AI insights panel, and an integrated chatbot trained on policy context."}],"whyItWorksHeading":"Why It Works: Outcome-Based Differentiation","whyItWorks":[{"title":"Speed to Value","description":"Most enterprise insurance software takes 6–12 months. We delivered a clickable AI prototype in 2 weeks and a production system in 8."},{"title":"AI-First Prototyping","description":"Every workflow was validated in the prototype before any production code, eliminating the costly 'build then redesign' loop. Pylon clicked through the actual product before we wrote a line of production code."},{"title":"Single Accountable Team","description":"Design, AI prototyping, and engineering under one roof - no handoff, no telephone game between agencies, no integration debt between teams."},{"title":"Outcome-Based Engagement","description":"Pylon paid for milestones and demonstrated results, not hours. Our incentives were tied directly to shipping production-quality software on schedule."}]},"es":{"client":"Pylon","clientLocation":"Estados Unidos","category":"InsurTech","duration":"Fase 1 (8 Semanas)","title":"Pylon - Operaciones de Seguros AI-Native, de Prototipo a Producción","tagline":"Plataforma de Operaciones de Seguros AI-Native","description":"Reemplazamos hojas de cálculo de pólizas fragmentadas y herramientas de siniestros desconectadas por un espacio de trabajo unificado impulsado por IA - diseñado y entregado en 8 semanas mediante prototipo clickable con IA.","scope":"Prototipado con IA, Diseño UI/UX, Desarrollo y Entrega","industry":"InsurTech","timeline":"8 Semanas","impacts":[{"value":"8 Semanas","label":"Prototipo a producción","description":"1 semana de alcance, 2 semanas de prototipo clickable con IA, 5 semanas de construcción."},{"value":"147K+","label":"Pólizas en una consola","description":"Cartera activa renderizada en un único espacio de trabajo con puntuación de riesgo en tiempo real."},{"value":"AI-Native","label":"Integrado en el flujo","description":"Insights de IA por póliza, detección de anomalías y un chatbot entrenado en contexto de pólizas."},{"value":"Equipo único","label":"Sin transferencias","description":"Prototipado con IA, diseño UI/UX e ingeniería entregados bajo un único equipo responsable."}],"problemShort":"El equipo de Pylon gestionaba miles de pólizas en hojas de cálculo, herramientas de siniestros separadas y bandejas de entrada - las señales de riesgo vivían en tres sistemas distintos y las renovaciones se rastreaban manualmente.","solutionShort":"Entregamos un espacio de trabajo AI-Native unificado - una consola para resumen de cartera, detalle de póliza, siniestros y renovaciones - validada mediante prototipo clickable con IA antes de cualquier código de producción.","challengesHeading":"El Desafío: Fragmentación a Escala","challengesIntro":"Un equipo de operaciones de seguros que gestiona miles de pólizas necesita una única fuente de verdad. Pylon no la tenía.","challenges":[{"title":"Flujo de Trabajo Fragmentado","description":"Pólizas, siniestros, renovaciones y riesgo vivían en herramientas separadas sin columna vertebral compartida. Los operadores reconciliaban estado entre sistemas a mano."},{"title":"Sin Visibilidad en Tiempo Real","description":"Las puntuaciones de riesgo y la urgencia de renovación se calculaban en lotes, dejando a los gestores reaccionar a paneles obsoletos en lugar de actuar sobre señales en vivo."},{"title":"Incorporación Lenta de Pólizas","description":"Añadir una nueva póliza significaba actualizar cuatro herramientas y una lista - horas por póliza y un rastro de auditoría que no cuadraba del todo."},{"title":"IA Sin Contexto","description":"El análisis existente eran informes estáticos. Un suscriptor no podía preguntar 'qué cambió en Henderson Corp la semana pasada' y obtener respuesta."}],"approachHeading":"Nuestro Enfoque: 3 Fases en 8 Semanas","approach":[{"number":"01","title":"Descubrimiento y Alcance","description":"Semana 1: nos integramos con suscriptores y operaciones para mapear cada flujo, clic y caso límite. Definimos el objetivo de producción y las métricas de éxito el día cinco."},{"number":"02","title":"Prototipo Clickable con IA","description":"Semanas 2–3: construimos un prototipo interactivo con herramientas de prototipado con IA para que el equipo de Pylon pudiera hacer clic por todo el producto - incluyendo insights de IA y respuestas del chatbot - y refinarlo en tiempo real antes de cualquier código de producción."},{"number":"03","title":"Construcción y Lanzamiento","description":"Semanas 4–8: entregamos la consola de producción con resumen de cartera, detalle por póliza, hub de siniestros, rastreador de renovaciones, panel de insights de IA y chatbot integrado entrenado en contexto de pólizas."}],"whyItWorksHeading":"Por Qué Funciona: Diferenciación Basada en Resultados","whyItWorks":[{"title":"Velocidad al Valor","description":"La mayoría del software empresarial tarda 6–12 meses. Entregamos un prototipo clickable con IA en 2 semanas y un sistema de producción en 8."},{"title":"Prototipado AI-First","description":"Cada flujo se validó en el prototipo antes de cualquier código de producción, eliminando el costoso ciclo 'construir y luego rediseñar'."},{"title":"Equipo Único Responsable","description":"Diseño, prototipado con IA e ingeniería bajo un mismo techo - sin transferencias, sin teléfono descompuesto entre agencias."},{"title":"Compromiso Basado en Resultados","description":"Pylon pagó por hitos y resultados demostrados, no por horas. Nuestros incentivos estaban directamente vinculados a entregar software de calidad de producción."}]},"fr":{"client":"Pylon","clientLocation":"États-Unis","category":"InsurTech","duration":"Phase 1 (8 Semaines)","title":"Pylon - Opérations d'Assurance AI-Native, du Prototype à la Production","tagline":"Plateforme d'Opérations d'Assurance AI-Native","description":"Remplacement de feuilles de calcul de polices fragmentées et d'outils de sinistres déconnectés par un espace de travail unifié piloté par l'IA - conçu et livré en 8 semaines via un prototype cliquable avec IA.","scope":"Prototypage IA, Design UI/UX, Développement et Livraison","industry":"InsurTech","timeline":"8 Semaines","impacts":[{"value":"8 Semaines","label":"Prototype à la production","description":"1 semaine de cadrage, 2 semaines de prototype IA cliquable, 5 semaines de production."},{"value":"147K+","label":"Polices dans une console","description":"Portefeuille actif rendu dans un seul espace de travail avec scoring de risque temps réel."},{"value":"AI-Native","label":"Intégré au workflow","description":"Insights IA par police, détection d'anomalies et chatbot entraîné sur le contexte des polices."},{"value":"Équipe unique","label":"Sans transfert","description":"Prototypage IA, design UI/UX et ingénierie livrés sous une seule équipe responsable."}],"problemShort":"L'équipe de Pylon gérait des milliers de polices entre tableurs, outils de sinistres séparés et boîtes mail - les signaux de risque vivaient dans trois systèmes différents et les renouvellements étaient suivis manuellement.","solutionShort":"Nous avons livré un espace de travail AI-Native unifié - une seule console pour vue de portefeuille, détail de police, sinistres et renouvellements - validé via prototype cliquable IA avant tout code de production.","challengesHeading":"Le Défi : Fragmentation à l'Échelle","challengesIntro":"Une équipe d'opérations d'assurance gérant des milliers de polices a besoin d'une source de vérité unique. Pylon n'en avait pas.","challenges":[{"title":"Workflow Fragmenté","description":"Polices, sinistres, renouvellements et risque vivaient dans des outils séparés sans colonne vertébrale partagée. Les opérateurs réconciliaient l'état entre systèmes à la main."},{"title":"Pas de Visibilité Temps Réel","description":"Les scores de risque et l'urgence des renouvellements étaient calculés en batch, laissant les gestionnaires réagir à des dashboards périmés."},{"title":"Onboarding de Police Lent","description":"Ajouter une nouvelle police signifiait mettre à jour quatre outils et une checklist - des heures par police et un audit trail incohérent."},{"title":"IA Sans Contexte","description":"Les analyses existantes étaient des rapports statiques. Un souscripteur ne pouvait pas demander 'qu'est-ce qui a changé chez Henderson Corp la semaine dernière' et obtenir une réponse."}],"approachHeading":"Notre Approche : 3 Phases en 8 Semaines","approach":[{"number":"01","title":"Découverte et Cadrage","description":"Semaine 1 : intégration avec souscripteurs et ops pour cartographier chaque flux, clic et cas limite. Cible de production et métriques définies au cinquième jour."},{"number":"02","title":"Prototype Cliquable IA","description":"Semaines 2–3 : prototype interactif avec outils de prototypage IA permettant à Pylon de cliquer dans tout le produit - y compris insights IA et chatbot - et de l'affiner en temps réel."},{"number":"03","title":"Build & Lancement","description":"Semaines 4–8 : console de production avec vue portefeuille, détail par police, hub de sinistres, suivi des renouvellements, panneau d'insights IA et chatbot intégré."}],"whyItWorksHeading":"Pourquoi Ça Marche : Différenciation par Résultats","whyItWorks":[{"title":"Vitesse de Valeur","description":"La plupart des logiciels d'entreprise prennent 6–12 mois. Nous avons livré un prototype IA cliquable en 2 semaines et un système de production en 8."},{"title":"Prototypage AI-First","description":"Chaque flux validé dans le prototype avant tout code de production, éliminant la coûteuse boucle 'construire puis redessiner'."},{"title":"Équipe Unique Responsable","description":"Design, prototypage IA et ingénierie sous un même toit - pas de transfert, pas de téléphone arabe entre agences."},{"title":"Engagement Basé sur les Résultats","description":"Pylon a payé pour des jalons et des résultats démontrés, pas pour des heures."}]},"de":{"client":"Pylon","clientLocation":"USA","category":"InsurTech","duration":"Phase 1 (8 Wochen)","title":"Pylon - AI-Native Versicherungsoperationen vom Prototyp zur Produktion","tagline":"AI-Native Versicherungsoperations-Plattform","description":"Ersatz fragmentierter Policen-Tabellen und getrennter Schadens-Tools durch einen einheitlichen, KI-gesteuerten Workspace - entworfen und ausgeliefert in 8 Wochen via klickbarem KI-Prototyp.","scope":"KI-Prototyping, UI/UX-Design, Entwicklung und Lieferung","industry":"InsurTech","timeline":"8 Wochen","impacts":[{"value":"8 Wochen","label":"Prototyp bis Produktion","description":"1 Woche Scoping, 2 Wochen klickbarer KI-Prototyp, 5 Wochen Production Build."},{"value":"147K+","label":"Policen in einer Konsole","description":"Aktives Portfolio in einem einzigen Workspace mit Echtzeit-Risikobewertung dargestellt."},{"value":"AI-Native","label":"Im Workflow eingebaut","description":"KI-Insights pro Police, Anomalie-Erkennung und Chatbot trainiert auf Police-Kontext."},{"value":"Ein Team","label":"Keine Übergaben","description":"KI-Prototyping, UI/UX-Design und Engineering unter einem verantwortlichen Team."}],"problemShort":"Das Pylon-Team verwaltete Tausende von Policen in Tabellen, separaten Schadenstools und Postfächern - Risikosignale in drei verschiedenen Systemen, Renewals manuell verfolgt.","solutionShort":"Wir lieferten einen einheitlichen AI-Native-Workspace - eine Konsole für Portfolio-Übersicht, Police-Detail, Schäden und Renewals - validiert via klickbarem KI-Prototyp vor jedem Production-Code.","challengesHeading":"Die Herausforderung: Fragmentierung im großen Maßstab","challengesIntro":"Ein Versicherungs-Ops-Team mit Tausenden von Policen braucht eine einzige Quelle der Wahrheit. Pylon hatte keine.","challenges":[{"title":"Fragmentierter Workflow","description":"Policen, Schäden, Renewals und Risiko lebten in separaten Tools ohne gemeinsamen Backbone. Bediener glichen Zustand zwischen Systemen manuell ab."},{"title":"Keine Echtzeit-Sichtbarkeit","description":"Risikoscores und Renewal-Dringlichkeit wurden in Batches berechnet, sodass Manager auf veraltete Dashboards reagierten."},{"title":"Langsames Police-Onboarding","description":"Eine neue Police hinzufügen bedeutete vier Tools und eine Checkliste zu aktualisieren - Stunden pro Police und ein nicht ganz konsistenter Audit-Trail."},{"title":"KI ohne Kontext","description":"Bestehende Analysen waren statische Reports. Ein Underwriter konnte nicht fragen 'was hat sich bei Henderson Corp letzte Woche geändert' und eine Antwort bekommen."}],"approachHeading":"Unser Ansatz: 3 Phasen in 8 Wochen","approach":[{"number":"01","title":"Discovery & Scoping","description":"Woche 1: eingebettet mit Underwritern und Ops, jeden Workflow, Klick und Edge-Case kartiert."},{"number":"02","title":"KI-Klickprototyp","description":"Wochen 2–3: interaktiver Prototyp mit KI-Prototyping-Tools, sodass Pylon durch das tatsächliche Produkt klicken konnte - inklusive KI-Insights und Chatbot."},{"number":"03","title":"Build & Launch","description":"Wochen 4–8: Produktionskonsole mit Portfolio-Übersicht, Police-Detail, Schaden-Hub, Renewal-Tracker, KI-Insights-Panel und Chatbot."}],"whyItWorksHeading":"Warum es funktioniert: Ergebnisbasierte Differenzierung","whyItWorks":[{"title":"Time-to-Value","description":"Die meisten Enterprise-Software-Projekte dauern 6–12 Monate. Wir lieferten einen klickbaren KI-Prototyp in 2 Wochen und ein Produktionssystem in 8."},{"title":"AI-First Prototyping","description":"Jeder Flow im Prototyp validiert vor jedem Production-Code - eliminiert die teure 'Bauen-dann-Neu-Designen'-Schleife."},{"title":"Ein verantwortliches Team","description":"Design, KI-Prototyping und Engineering unter einem Dach - keine Übergaben, keine stille Post zwischen Agenturen."},{"title":"Ergebnisbasiertes Engagement","description":"Pylon zahlte für Meilensteine und nachweisbare Ergebnisse, nicht für Stunden."}]},"ru":{"client":"Pylon","clientLocation":"США","category":"InsurTech","duration":"Фаза 1 (8 Недель)","title":"Pylon - AI-Native Страховые Операции от Прототипа до Продакшена","tagline":"AI-Native Платформа Страховых Операций","description":"Заменили фрагментированные таблицы полисов и разрозненные инструменты убытков единым рабочим пространством с ИИ - спроектировано и запущено за 8 недель через кликабельный AI-прототип.","scope":"AI-прототипирование, UI/UX-дизайн, Разработка и Внедрение","industry":"InsurTech","timeline":"8 Недель","impacts":[{"value":"8 Недель","label":"Прототип в продакшен","description":"1 неделя scope, 2 недели кликабельного AI-прототипа, 5 недель сборки."},{"value":"147K+","label":"Полисов в одной консоли","description":"Активный портфель в едином рабочем пространстве с риск-скорингом в реальном времени."},{"value":"AI-Native","label":"Встроено в рабочий процесс","description":"AI-инсайты по полису, обнаружение аномалий и чатбот, обученный на контексте полисов."},{"value":"Одна команда","label":"Без передач","description":"AI-прототипирование, UI/UX-дизайн и инжиниринг от единой ответственной команды."}],"problemShort":"Команда Pylon управляла тысячами полисов в таблицах, отдельных инструментах убытков и почтовых ящиках - сигналы риска жили в трёх разных системах, продления отслеживались вручную.","solutionShort":"Мы поставили единое AI-Native рабочее пространство - одну консоль для обзора портфеля, деталей полиса, убытков и продлений - проверено через кликабельный AI-прототип до любого продакшен-кода.","challengesHeading":"Вызов: Фрагментация в Масштабе","challengesIntro":"Команде страховых операций с тысячами полисов нужен единый источник истины. У Pylon его не было.","challenges":[{"title":"Фрагментированный Рабочий Процесс","description":"Полисы, убытки, продления и риск жили в разных инструментах без общего стержня."},{"title":"Нет Видимости в Реальном Времени","description":"Риск-скоры и срочность продлений считались пакетно, оставляя менеджеров реагировать на устаревшие дашборды."},{"title":"Медленный Онбординг Полиса","description":"Добавление новой клиентской полиса означало обновление четырёх инструментов и чек-листа - часы на полис."},{"title":"ИИ Без Контекста","description":"Существующая аналитика была статичными отчётами. Андеррайтер не мог спросить 'что изменилось у Henderson Corp на прошлой неделе'."}],"approachHeading":"Наш Подход: 3 Фазы за 8 Недель","approach":[{"number":"01","title":"Discovery и Scoping","description":"Неделя 1: погружение с андеррайтерами и операторами для отрисовки каждого потока, клика и крайнего случая."},{"number":"02","title":"Кликабельный AI-Прототип","description":"Недели 2–3: интерактивный прототип с AI-инструментами, чтобы команда Pylon могла кликать по реальному продукту."},{"number":"03","title":"Сборка и Запуск","description":"Недели 4–8: продакшен-консоль с обзором портфеля, деталями полиса, хабом убытков, трекером продлений и интегрированным чатботом."}],"whyItWorksHeading":"Почему Это Работает: Дифференциация по Результатам","whyItWorks":[{"title":"Скорость к Ценности","description":"Большинство корпоративного ПО занимает 6–12 месяцев. Мы поставили AI-прототип за 2 недели и продакшен-систему за 8."},{"title":"AI-First Прототипирование","description":"Каждый поток был проверен в прототипе до продакшен-кода."},{"title":"Одна Ответственная Команда","description":"Дизайн, AI-прототипирование и инжиниринг под одной крышей."},{"title":"Ангажемент по Результатам","description":"Pylon платил за вехи и продемонстрированные результаты, а не за часы."}]},"nl":{"client":"Pylon","clientLocation":"Verenigde Staten","category":"InsurTech","duration":"Fase 1 (8 Weken)","title":"Pylon - AI-Native Verzekeringsoperaties van Prototype naar Productie","tagline":"AI-Native Verzekeringsoperaties Platform","description":"Vervangen van gefragmenteerde polis-spreadsheets en losgekoppelde claim-tools door een uniforme AI-gedreven werkruimte - ontworpen en geleverd in 8 weken via klikbaar AI-prototype.","scope":"AI-Prototyping, UI/UX-Ontwerp, Ontwikkeling en Levering","industry":"InsurTech","timeline":"8 Weken","impacts":[{"value":"8 Weken","label":"Prototype tot productie","description":"1 week scope, 2 weken klikbaar AI-prototype, 5 weken productie-build."},{"value":"147K+","label":"Polissen in één console","description":"Actieve portefeuille gerenderd in een enkele werkruimte met real-time risk scoring."},{"value":"AI-Native","label":"Ingebouwd in workflow","description":"AI-inzichten per polis, anomaliedetectie en chatbot getraind op poliscontext."},{"value":"Eén team","label":"Geen overdrachten","description":"AI-prototyping, UI/UX-ontwerp en engineering geleverd onder één verantwoordelijk team."}],"problemShort":"Het Pylon-team beheerde duizenden polissen verspreid over spreadsheets, losse claim-tools en mailboxen - risicosignalen leefden in drie verschillende systemen.","solutionShort":"We leverden een uniforme AI-Native werkruimte - één console voor portefeuille-overzicht, polisdetail, claims en verlengingen - gevalideerd via klikbaar AI-prototype voor enige productiecode.","challengesHeading":"De Uitdaging: Fragmentatie op Schaal","challengesIntro":"Een verzekeringsteam met duizenden polissen heeft één bron van waarheid nodig. Pylon had die niet.","challenges":[{"title":"Gefragmenteerde Workflow","description":"Polissen, claims, verlengingen en risico leefden in losse tools zonder gedeelde ruggengraat."},{"title":"Geen Real-Time Zicht","description":"Risicoscores en verlengingsurgentie werden in batches berekend, waardoor managers reageerden op verouderde dashboards."},{"title":"Trage Polis Onboarding","description":"Een nieuwe polis toevoegen betekende vier tools en een checklist bijwerken - uren per polis."},{"title":"AI Zonder Context","description":"Bestaande analyses waren statische rapporten. Een underwriter kon niet vragen 'wat veranderde bij Henderson Corp vorige week'."}],"approachHeading":"Onze Aanpak: 3 Fases in 8 Weken","approach":[{"number":"01","title":"Discovery & Scoping","description":"Week 1: ingebed bij underwriters en ops om elke flow, klik en edge-case te mappen."},{"number":"02","title":"Klikbaar AI-Prototype","description":"Weken 2–3: interactief prototype zodat het Pylon-team door het echte product kon klikken inclusief AI-inzichten en chatbot."},{"number":"03","title":"Build & Lancering","description":"Weken 4–8: productieconsole met portefeuille-overzicht, polisdetail, claim-hub, verlengingstracker en geïntegreerde chatbot."}],"whyItWorksHeading":"Waarom Het Werkt: Resultaatgerichte Differentiatie","whyItWorks":[{"title":"Snelheid naar Waarde","description":"De meeste enterprise-software duurt 6–12 maanden. Wij leverden een klikbaar AI-prototype in 2 weken en een productiesysteem in 8."},{"title":"AI-First Prototyping","description":"Elke flow gevalideerd in het prototype voor enige productiecode."},{"title":"Eén Verantwoordelijk Team","description":"Ontwerp, AI-prototyping en engineering onder één dak."},{"title":"Resultaatgericht Engagement","description":"Pylon betaalde voor mijlpalen en aangetoonde resultaten, niet uren."}]},"ja":{"client":"Pylon","clientLocation":"アメリカ","category":"InsurTech","duration":"フェーズ1(8週間)","title":"Pylon - AI-Nativeな保険業務、プロトタイプから本番稼働まで","tagline":"AI-Native保険業務プラットフォーム","description":"断片化していた保険証券スプレッドシートと切り離された請求ツールを、統合されたAI駆動のワークスペースに置き換え - 8週間でクリック可能なAIプロトタイプを通じて設計・納品。","scope":"AIプロトタイピング、UI/UXデザイン、開発、納品","industry":"InsurTech","timeline":"8週間","impacts":[{"value":"8週間","label":"プロトタイプから本番へ","description":"1週間スコープ、2週間クリック可能AIプロトタイプ、5週間本番構築。"},{"value":"147K+","label":"1コンソールで全証券","description":"アクティブなポートフォリオを単一ワークスペースでリアルタイムリスクスコアリング付きで表示。"},{"value":"AI-Native","label":"ワークフローに組込み","description":"証券ごとのAIインサイト、異常検知、証券コンテキストで訓練されたチャットボット。"},{"value":"単一チーム","label":"引継ぎなし","description":"AIプロトタイピング、UI/UXデザイン、エンジニアリングを単一の責任チームで納品。"}],"problemShort":"Pylonのチームは数千の証券をスプレッドシート、別個の請求ツール、メールボックス間で管理していた - リスクシグナルは3つの異なるシステムに散在。","solutionShort":"ポートフォリオ概要、証券詳細、請求、更新の単一コンソール - 本番コードの前にクリック可能なAIプロトタイプで検証。","challengesHeading":"課題:スケールにおけるフラグメンテーション","challengesIntro":"数千の証券を扱う保険業務チームには単一の真実の源が必要。Pylonにはなかった。","challenges":[{"title":"断片化したワークフロー","description":"証券、請求、更新、リスクが共通の背骨なしに別々のツールに存在。"},{"title":"リアルタイム可視性の欠如","description":"リスクスコアと更新の緊急度はバッチ処理で計算され、マネージャーは古いダッシュボードに反応していた。"},{"title":"遅い証券オンボーディング","description":"新しい証券の追加には4つのツールとチェックリストの更新が必要 - 証券あたり数時間。"},{"title":"コンテキストのないAI","description":"既存の分析は静的レポート。アンダーライターが「先週Henderson Corpで何が変わった?」と質問しても答えは得られなかった。"}],"approachHeading":"アプローチ:8週間で3フェーズ","approach":[{"number":"01","title":"ディスカバリーとスコーピング","description":"1週目:アンダーライターと運用に組み込み、すべてのフロー、クリック、エッジケースをマッピング。"},{"number":"02","title":"AI駆動クリック可能プロトタイプ","description":"2-3週目:Pylonチームが実際の製品をクリックスルーできるインタラクティブプロトタイプ。"},{"number":"03","title":"本番構築と立ち上げ","description":"4-8週目:ポートフォリオ概要、証券詳細、請求ハブ、更新トラッカー、AIインサイトパネル、統合チャットボット付き本番コンソール。"}],"whyItWorksHeading":"成功理由:結果ベースの差別化","whyItWorks":[{"title":"価値実現の速さ","description":"ほとんどのエンタープライズソフトウェアは6-12ヶ月。クリック可能AIプロトタイプを2週間、本番システムを8週間で納品。"},{"title":"AI-Firstプロトタイピング","description":"本番コードの前に各フローをプロトタイプで検証 - コストのかかる「構築してから再設計」ループを排除。"},{"title":"単一の責任チーム","description":"デザイン、AIプロトタイピング、エンジニアリングを一つ屋根の下で - 引継ぎなし、エージェンシー間の伝言ゲームなし。"},{"title":"結果ベースのエンゲージメント","description":"Pylonはマイルストーンと実証された結果に対して支払い、時間ではない。"}]},"ko":{"client":"Pylon","clientLocation":"미국","category":"InsurTech","duration":"1단계 (8주)","title":"Pylon - 프로토타입에서 프로덕션까지의 AI-Native 보험 운영","tagline":"AI-Native 보험 운영 플랫폼","description":"분산된 보험 증권 스프레드시트와 분리된 클레임 도구를 통합 AI 기반 워크스페이스로 교체 - 클릭 가능한 AI 프로토타입을 통해 8주 만에 설계 및 출시.","scope":"AI 프로토타이핑, UI/UX 디자인, 개발 및 납품","industry":"InsurTech","timeline":"8주","impacts":[{"value":"8주","label":"프로토타입에서 프로덕션","description":"1주 범위, 2주 클릭 가능 AI 프로토타입, 5주 프로덕션 빌드."},{"value":"147K+","label":"한 콘솔에서 모든 증권","description":"활성 포트폴리오를 단일 워크스페이스에서 실시간 리스크 스코어링과 함께 렌더링."},{"value":"AI-Native","label":"워크플로에 내장","description":"증권별 AI 인사이트, 이상 감지, 증권 컨텍스트로 학습된 챗봇."},{"value":"단일 팀","label":"핸드오프 없음","description":"AI 프로토타이핑, UI/UX 디자인, 엔지니어링을 단일 책임 팀으로 납품."}],"problemShort":"Pylon 팀은 수천 개의 증권을 스프레드시트, 별도 클레임 도구, 받은편지함에서 관리 - 리스크 신호가 세 가지 다른 시스템에 존재.","solutionShort":"포트폴리오 개요, 증권 상세, 클레임, 갱신을 위한 단일 콘솔 - 프로덕션 코드 전 클릭 가능 AI 프로토타입으로 검증.","challengesHeading":"도전 과제: 규모에서의 분절화","challengesIntro":"수천 개의 증권을 운영하는 보험 운영 팀은 단일 진실의 원천이 필요하다. Pylon에는 없었다.","challenges":[{"title":"분절된 워크플로","description":"증권, 클레임, 갱신, 리스크가 공유된 척추 없이 분리된 도구에 존재."},{"title":"실시간 가시성 없음","description":"리스크 점수와 갱신 긴급도가 배치로 계산되어 매니저들이 오래된 대시보드에 반응."},{"title":"느린 증권 온보딩","description":"새 증권 추가는 4개의 도구와 체크리스트 업데이트 필요 - 증권당 몇 시간."},{"title":"컨텍스트 없는 AI","description":"기존 분석은 정적 보고서. 인수인이 '지난주 Henderson Corp에서 무엇이 바뀌었나'를 물을 수 없었다."}],"approachHeading":"접근 방식: 8주 동안 3단계","approach":[{"number":"01","title":"디스커버리 및 스코핑","description":"1주차: 인수인과 운영팀에 내장되어 모든 흐름, 클릭, 엣지 케이스 매핑."},{"number":"02","title":"AI 기반 클릭 가능 프로토타입","description":"2-3주차: Pylon 팀이 실제 제품을 클릭할 수 있는 인터랙티브 프로토타입."},{"number":"03","title":"프로덕션 빌드 및 출시","description":"4-8주차: 포트폴리오 개요, 증권 상세, 클레임 허브, 갱신 트래커, AI 인사이트 패널, 통합 챗봇 포함 프로덕션 콘솔."}],"whyItWorksHeading":"작동 이유: 결과 기반 차별화","whyItWorks":[{"title":"가치 실현 속도","description":"대부분의 엔터프라이즈 소프트웨어는 6-12개월. 클릭 가능 AI 프로토타입을 2주, 프로덕션 시스템을 8주에 납품."},{"title":"AI-First 프로토타이핑","description":"프로덕션 코드 전 모든 흐름을 프로토타입에서 검증."},{"title":"단일 책임 팀","description":"디자인, AI 프로토타이핑, 엔지니어링을 한 지붕 아래에서."},{"title":"결과 기반 계약","description":"Pylon은 마일스톤과 입증된 결과에 대해 지불, 시간이 아닌."}]},"el":{"client":"Pylon","clientLocation":"ΗΠΑ","category":"InsurTech","duration":"Φάση 1 (8 Εβδομάδες)","title":"Pylon - AI-Native Ασφαλιστικές Λειτουργίες από το Πρωτότυπο στην Παραγωγή","tagline":"AI-Native Πλατφόρμα Ασφαλιστικών Λειτουργιών","description":"Αντικαταστάθηκαν τα κατακερματισμένα υπολογιστικά φύλλα ασφαλιστηρίων και τα αποσυνδεδεμένα εργαλεία αξιώσεων από έναν ενοποιημένο χώρο εργασίας με AI - σχεδιάστηκε και παραδόθηκε σε 8 εβδομάδες μέσω κλικαβλέ AI-πρωτοτύπου.","scope":"AI Πρωτοτύπωση, UI/UX Σχεδιασμός, Ανάπτυξη και Παράδοση","industry":"InsurTech","timeline":"8 Εβδομάδες","impacts":[{"value":"8 Εβδομάδες","label":"Πρωτότυπο προς παραγωγή","description":"1 εβδομάδα scope, 2 εβδομάδες κλικαβλέ AI-πρωτότυπο, 5 εβδομάδες παραγωγή."},{"value":"147K+","label":"Ασφαλιστήρια σε μία κονσόλα","description":"Ενεργό χαρτοφυλάκιο σε ενιαίο χώρο εργασίας με βαθμολόγηση κινδύνου σε πραγματικό χρόνο."},{"value":"AI-Native","label":"Ενσωματωμένο στη ροή","description":"AI insights ανά ασφαλιστήριο, ανίχνευση ανωμαλιών και chatbot εκπαιδευμένο σε context ασφαλιστηρίων."},{"value":"Ενιαία ομάδα","label":"Χωρίς παραδώσεις","description":"AI πρωτοτύπωση, UI/UX σχεδιασμός και μηχανική σε μία υπεύθυνη ομάδα."}],"problemShort":"Η ομάδα της Pylon διαχειριζόταν χιλιάδες ασφαλιστήρια σε υπολογιστικά φύλλα, ξεχωριστά εργαλεία αξιώσεων και inboxes - τα σήματα κινδύνου ζούσαν σε τρία διαφορετικά συστήματα.","solutionShort":"Παραδώσαμε έναν ενοποιημένο AI-Native χώρο εργασίας - μία κονσόλα για επισκόπηση χαρτοφυλακίου, λεπτομέρειες ασφαλιστηρίου, αξιώσεις και ανανεώσεις - επικυρωμένη μέσω κλικαβλέ AI-πρωτοτύπου.","challengesHeading":"Η Πρόκληση: Κατακερματισμός σε Κλίμακα","challengesIntro":"Μια ομάδα ασφαλιστικών λειτουργιών με χιλιάδες ασφαλιστήρια χρειάζεται μία ενιαία πηγή αλήθειας. Η Pylon δεν είχε.","challenges":[{"title":"Κατακερματισμένη Ροή Εργασίας","description":"Ασφαλιστήρια, αξιώσεις, ανανεώσεις και κίνδυνος ζούσαν σε ξεχωριστά εργαλεία χωρίς κοινή ραχοκοκαλιά."},{"title":"Καμία Ορατότητα σε Πραγματικό Χρόνο","description":"Οι βαθμοί κινδύνου και η επείγουσα ανάγκη ανανέωσης υπολογίζονταν σε batches."},{"title":"Αργή Ένταξη Ασφαλιστηρίου","description":"Η προσθήκη ενός νέου ασφαλιστηρίου σήμαινε ενημέρωση τεσσάρων εργαλείων."},{"title":"AI Χωρίς Πλαίσιο","description":"Οι υπάρχουσες αναλύσεις ήταν στατικές αναφορές."}],"approachHeading":"Η Προσέγγισή μας: 3 Φάσεις σε 8 Εβδομάδες","approach":[{"number":"01","title":"Ανακάλυψη και Scoping","description":"Εβδομάδα 1: ενσωματωμένοι με ασφαλιστές και ops για να χαρτογραφήσουμε κάθε ροή."},{"number":"02","title":"AI Κλικαβλέ Πρωτότυπο","description":"Εβδομάδες 2-3: διαδραστικό πρωτότυπο που η ομάδα της Pylon μπορούσε να κάνει κλικ μέσα από το πραγματικό προϊόν."},{"number":"03","title":"Παραγωγή και Εκκίνηση","description":"Εβδομάδες 4-8: κονσόλα παραγωγής με επισκόπηση χαρτοφυλακίου, λεπτομέρειες ασφαλιστηρίου, hub αξιώσεων, tracker ανανεώσεων και chatbot."}],"whyItWorksHeading":"Γιατί Λειτουργεί: Διαφοροποίηση Βάσει Αποτελεσμάτων","whyItWorks":[{"title":"Ταχύτητα στην Αξία","description":"Το περισσότερο εταιρικό λογισμικό παίρνει 6-12 μήνες. Παραδώσαμε κλικαβλέ AI-πρωτότυπο σε 2 εβδομάδες και σύστημα παραγωγής σε 8."},{"title":"AI-First Πρωτοτύπωση","description":"Κάθε ροή επικυρώθηκε στο πρωτότυπο πριν από οποιονδήποτε κώδικα παραγωγής."},{"title":"Μία Υπεύθυνη Ομάδα","description":"Σχεδιασμός, AI πρωτοτύπωση και μηχανική κάτω από μία στέγη."},{"title":"Δέσμευση Βάσει Αποτελεσμάτων","description":"Η Pylon πλήρωσε για ορόσημα και αποδεδειγμένα αποτελέσματα."}]}},"others":[{"meta":{"slug":"kittiwake","industry":"insurance","projectType":"AI claims intake","country":"gb","tags":["insurtech","ai","enterprise"],"compliances":["FCA ICOBS 8","FCA Consumer Duty","FCA FG21/1","CIDRA 2012","Insurance Act 2015","UK GDPR Article 22","Data (Use and Access) Act 2025","EU AI Act","FCA DISP"],"cover":{"src":"/case-studies/kittiwake/01-home.png","width":1440,"height":1000}},"text":{"en":{"title":"Kittiwake - AI Claims Intake with a Person Behind Every Decision","description":"Designed and built a clickable claims intake console for a UK motor and home insurer: AI reads the customer's story and photos, extracts the facts, drafts a damage estimate and suggests a triage route with its reasons, while a named person confirms cover, routes the claim and makes every call that could go against the customer.","category":"Insurance / AI Claims Intake","client":"Brackenmoor Insurance"}}},{"meta":{"slug":"gratwerk","industry":"insurance","projectType":"Underwriting workbench","country":"ch","tags":["insurtech","enterprise","ai"],"compliances":["ISA / VAG and AVO","VVG / LCA","revFADP / nDSG","Embargo Act (SECO sanctions)","Federal stamp duty","Natural hazard uniform tariff","Cantonal building insurance"],"cover":{"src":"/case-studies/gratwerk/01-home.png","width":1440,"height":1000}},"text":{"en":{"title":"Gratwerk - Commercial Underwriting Workbench from Broker Submission to Quote","description":"Designed and built a clickable underwriting workbench for a Bern-based insurer writing commercial property and liability through brokers: a submission inbox with extraction already done, an underwriter queue with the quote SLA, a submission file enriched with hazard, building and company data, and a workbench that rates the risk, checks every referral rule and shows who may quote.","category":"Insurance / Underwriting Workbench","client":"Aarewald Versicherungen AG"}}},{"meta":{"slug":"ormeggio","industry":"insurance","projectType":"Policy administration","country":"it","tags":["insurtech","enterprise"],"compliances":["Codice delle Assicurazioni Private","IVASS Regulation 40/2018","IVASS Regulation 41/2018","EU IPID","SEPA Core Direct Debit","GDPR","Italian Privacy Code"],"cover":{"src":"/case-studies/ormeggio/01-home.png","width":1440,"height":1000}},"text":{"en":{"title":"Ormeggio - Policy Administration for Italian Motor and Home Insurance","description":"Designed and built a clickable policy administration console for an Italian motor and home insurer: every policy with its versions, coverages, taxes and bonus-malus class, and an endorsement workbench that works out the pro-rata premium per coverage and checks the approver's authority before a change goes through.","category":"Insurance / Policy Administration","client":"Corvalta Assicurazioni"}}},{"meta":{"slug":"kemudi","industry":"insurance","projectType":"Broker & agent portal","country":"my","tags":["insurtech","enterprise"],"compliances":["Financial Services Act 2013","BNM Fair Treatment of Financial Consumers","BNM Product Transparency and Disclosure","BNM e-KYC","AMLA 2001","PDPA 2010","Road Transport Act 1987","Electronic Commerce Act 2006","Form CP58 (Income Tax Act 1967)"],"cover":{"src":"/case-studies/kemudi/01-home.png","width":1440,"height":1000}},"text":{"en":{"title":"Kemudi - Broker and Agent Portal for Quotes, Applications and Commission","description":"Designed and built a clickable portal for the agents and brokers of a Malaysian insurance group with a life and a general insurer: a daily to-do list, quick quotes gated by the agent's licence, every case from quote to issued policy, application tracking through underwriting requirements, and commission statements with clawbacks shown before they happen.","category":"Insurance / Broker & Agent Portal","client":"Seri Andalas"}}},{"meta":{"slug":"tussock","industry":"insurance","projectType":"Customer self-service app","country":"nz","tags":["insurtech","mobile"],"compliances":["CoFI Act 2022","Insurance (Prudential Supervision) Act 2010","Contracts of Insurance Act 2024","Fair Trading Act 1986","Natural Hazards Insurance Act 2023","Privacy Act 2020","Fair Insurance Code","WCAG 2.2 AA"],"cover":{"src":"/case-studies/tussock/01-home.png","width":1440,"height":1000}},"text":{"en":{"title":"Tussock - Customer Self-Service App for Home, Contents, Car and Travel Insurance","description":"Designed and built a clickable self-service app for a New Zealand personal lines insurer: every policy on one home screen, cover in standard or plain language, renewals that explain each reason for a price change, claims with Natural Hazards Commission steps on the timeline, and payments with failed instalments fixed in a tap.","category":"Insurance / Customer Self-Service","client":"Kōtuku Insurance"}}}]}