<?xml version='1.0' encoding='utf-8' ?>
<!-- Made with love by pretalx v2024.3.1. -->
<schedule>
    <generator name="pretalx" version="2024.3.1" />
    <version>0.2</version>
    <conference>
        <title>Software Crafters Barcelona 2026</title>
        <acronym>scbcn2026</acronym>
        <start>2026-10-16</start>
        <end>2026-10-17</end>
        <days>2</days>
        <timeslot_duration>00:05</timeslot_duration>
        <base_url>https://c4p.softwarecrafters.barcelona</base_url>
        <logo>https://c4p.softwarecrafters.barcelona/media/scbcn2026/img/Software_Crafters_pretalx_Vv2MKl2_GIDdBFt.png</logo>
        <time_zone_name>Europe/Madrid</time_zone_name>
        
        
        <track name="Innovation and Trends" slug="12-innovation-and-trends"  color="#1976d2" />
        
        <track name="Development principles and methodologies" slug="13-development-principles-and-methodologies"  color="#388e3c" />
        
        <track name="Software Design and Architecture" slug="14-software-design-and-architecture"  color="#f57c00" />
        
        <track name="Testing and Quality" slug="15-testing-and-quality"  color="#00897b" />
        
        <track name="Diversity in IT" slug="16-diversity-in-it"  color="#7b1fa2" />
        
        <track name="AI and Automation" slug="17-ai-and-automation"  color="#d32f2f" />
        
        <track name="Other" slug="19-other"  color="#bc6cc0" />
        
        <track name="Monitoring and observability" slug="20-monitoring-and-observability"  color="#00ccff" />
        
        <track name="Tech culture and soft skills" slug="21-tech-culture-and-soft-skills"  color="#33ff00" />
        
        <track name="Accesibility and sustainability" slug="22-accesibility-and-sustainability"  color="#fff600" />
        
    </conference>
    <day index='1' date='2026-10-16' start='2026-10-16T04:00:00+02:00' end='2026-10-17T03:59:00+02:00'>
        
    </day>
    <day index='2' date='2026-10-17' start='2026-10-17T04:00:00+02:00' end='2026-10-18T03:59:00+02:00'>
        <room name='Track 1' guid='2a45fd16-73b0-5f89-a5c7-e0f4906e73ac'>
            <event guid='dc6f14e3-eeba-5f0c-a273-6d792c5143db' id='230'>
                <room>Track 1</room>
                <title>Next.js: The Final Boss of Web Performance</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T10:00:00+02:00</date>
                <start>10:00</start>
                <duration>00:55</duration>
                <abstract>Your local Lighthouse test shows a perfect 100, but when your app hits the real-world production stage, it&#8217;s instant Game Over. Welcome to the Next.js Hard Mode!

In this data-driven session, we will bypass marketing hype and put Next.js&apos;s real-world performance to the test. Armed with massive global datasets from the HTTP Archive and the Chrome User Experience Report (CrUX), we&#8217;ll review the built-in optimization power-ups Next.js provides, contrast them with the messy reality of large-scale projects, and hand you a practical cheat code survival kit to rescue your Core Web Vitals before you run out of lives.</abstract>
                <slug>scbcn2026-230-next-js-the-final-boss-of-web-performance</slug>
                <track>Development principles and methodologies</track>
                
                <persons>
                    <person id='178'>Estela Franco</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/CDWLGX/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/CDWLGX/feedback/</feedback_url>
            </event>
            <event guid='e3757573-188e-5697-9712-53709fc3182f' id='195'>
                <room>Track 1</room>
                <title>Construint un framework de testing distribu&#239;t amb Argo Workflows i Open Test Reporting</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T11:30:00+02:00</date>
                <start>11:30</start>
                <duration>00:55</duration>
                <abstract>La nostra plataforma basada en Kafka gestiona milions de connexions, amb el suport de desenes de components en Kubernetes. La validaci&#243; autom&#224;tica i el testing funcional s&#243;n essencials per al nostre proc&#233;s de release. El framework de testing que vam construir originalment ens va permetre executar tests end-to-end durant anys, per&#242;, a mesura que la plataforma es va anar ampliant, van anar sorgint diversos problemes: temps d&#8217;execuci&#243; llargs, integraci&#243; limitada amb Kubernetes, tooling fragmentat i una fricci&#243; creixent tant per als developers com per a QA.

Durant l&#8217;&#250;ltim any, hem estat modernitzant aquest legacy framework mentre continu&#224;vem amb les releases de la plataforma sota terminis ajustats. En aquesta xerrada compartirem com vam abordar aquest proc&#233;s: identificant els principals pain points, definint els building blocks d&#8217;una nova soluci&#243; i introduint de manera incremental tecnologies com Argo Workflows, deployments amb Helm i Flux, un nou ecosistema de components I reporting estandarditzat.

Juntament amb el recorregut t&#232;cnic, tamb&#233; parlarem de les consideracions pr&#224;ctiques que han marcat el projecte &#8212; equilibrar la migraci&#243; amb el desenvolupament continu, coordinar els esfor&#231;os i introduir millores sense poder &#8220;aturar-ho tot&#8221;. Els assistents s&#8217;enduran estrat&#232;gies per fer evolucionar un sistema de testing complex en Kubernetes, aix&#237; com aprenentatges tant des del vessant t&#232;cnic com des de l&#8217;organitzatiu a l&#8217;hora de modernitzar una pe&#231;a cr&#237;tica d&apos;infraestuctura.</abstract>
                <slug>scbcn2026-195-construint-un-framework-de-testing-distribut-amb-argo-workflows-i-open-test-reporting</slug>
                <track>Testing and Quality</track>
                
                <persons>
                    <person id='144'>Javier Rois</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>true</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/3KV8NQ/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/3KV8NQ/feedback/</feedback_url>
            </event>
            <event guid='d6db124c-1b44-5374-9236-38bbb18e2c56' id='141'>
                <room>Track 1</room>
                <title>De 300 ms a 3 segundos: C&#243;mo navegar la tormenta de un incidente real  en producci&#243;n y salir airoso</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T12:30:00+02:00</date>
                <start>12:30</start>
                <duration>00:55</duration>
                <abstract>Parec&#237;a un d&#237;a normal: las alertas habituales, el ruido esperable en un sistema distribuido como Grafana Tempo, nuestra base de datos de trazas. Hasta que lleg&#243; una de las gordas, y yo estaba de guardia.

El p99 de la latencia de escritura en uno de nuestros cl&#250;steres m&#225;s grandes pas&#243; de cientos de milisegundos a m&#225;s de 3 segundos. En un cl&#250;ster que ingiere 64 TB al d&#237;a, esto no es anecd&#243;tico. Los timeouts empezaron a multiplicarse y, con ellos, la incertidumbre sobre millones de spans potencialmente perdidos. Declar&#233; el incidente.

Esta charla trata de lo que pasa cuando todo se rompe en producci&#243;n y t&#250; est&#225;s al mando. A partir de este incidente real, veremos c&#243;mo convertir una crisis en una oportunidad de crecimiento t&#233;cnico y profesional.

&#191;Qu&#233; vas a aprender?
* Triaje sin p&#225;nico: C&#243;mo responder durante los primeros minutos cr&#237;ticos, definir roles (Comandante e Investigador) y comunicar con claridad y transparencia cuando la causa no est&#225; clara.
* Diagn&#243;stico pragm&#225;tico: T&#233;cnicas para evitar el sesgo de confirmaci&#243;n y la visi&#243;n de t&#250;nel cuando las alertas no dejan de sonar.
* Post-mortems que importan: C&#243;mo escribir an&#225;lisis blameless (sin culpas) que generen cambios sist&#233;micos, no documentos olvidados.
* Crecimiento profesional: Por qu&#233; los incidentes son uno de los aceleradores m&#225;s poderosos en la carrera de un ingeniero.

Cerraremos con ejercicio simulado, poniendo en pr&#225;ctica las las herramientas de triaje y comunicaci&#243;n aprendidas.</abstract>
                <slug>scbcn2026-141-de-300-ms-a-3-segundos-cmo-navegar-la-tormenta-de-un-incidente-real-en-produccin-y-salir-airoso</slug>
                <track>Monitoring and observability</track>
                
                <persons>
                    <person id='116'>Javi</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/M9TDG9/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/M9TDG9/feedback/</feedback_url>
            </event>
            <event guid='05c5b840-47b1-5d7a-bcbc-d23078d062f4' id='209'>
                <room>Track 1</room>
                <title>&#191;Es rentable (y posible) la IA 100% local en desarrollo?</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T15:00:00+02:00</date>
                <start>15:00</start>
                <duration>00:55</duration>
                <abstract>Tal y como est&#225; el patio de las suscripciones de Claude, Antigravity, Codex.... &#191;Es factible hoy en d&#237;a desarrollar software profesional prescindiendo de los modelos privativos y sus sistemas de pago por uso?

Movido por la b&#250;squeda de autonom&#237;a t&#233;cnica y privacidad, decid&#237; realizar una inversi&#243;n &quot;posiblemente descabellada&quot; en hardware de &#250;ltima generaci&#243;n (NVIDIA Blackwell) para ejecutar modelos de pesos abiertos y ponerlo a prueba.

Esta es la cr&#243;nica de un descubrimiento: un viaje de aprendizaje incremental donde se analiza, desde una visi&#243;n cr&#237;tica y honesta, si la IA local ha madurado lo suficiente para el d&#237;a a d&#237;a de un desarrollador. 

Exploraremos c&#243;mo configurar un nodo de c&#243;mputo, el dilema de las cuantizaciones, qu&#233; modelos elegir para cada fase del ciclo de vida (planificar, programar, refactorizar) y c&#243;mo se comportan estas herramientas ante retos reales como el Domain-Driven Design (DDD) sobre un producto complejo.

Una sesi&#243;n para quienes se preguntan si es el momento de reclamar su soberan&#237;a tecnol&#243;gica o si todav&#237;a estamos atados a la nube.</abstract>
                <slug>scbcn2026-209--es-rentable-y-posible-la-ia-100-local-en-desarrollo-</slug>
                <track>AI and Automation</track>
                
                <persons>
                    <person id='2'>Fernando Aparicio Reviriego</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/LUPUHA/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/LUPUHA/feedback/</feedback_url>
            </event>
            <event guid='d294b986-27d3-5589-b0cf-e3701c947a9a' id='226'>
                <room>Track 1</room>
                <title>Testing the Limits: Can PMs and Designers Build Products in the Real World?</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T16:00:00+02:00</date>
                <start>16:00</start>
                <duration>00:55</duration>
                <abstract>With all the recent advancements in AI, we found ourselves, almost inevitably, uncovering better ways of developing software by doing it and helping others do it. This led us to a fundamental question: Can a pair of Product Managers and Designers build something real, reliable, and production-ready with minimal developer supervision? We wanted to move past the marketing hype and find out if non-engineers could actually navigate the complexities of a live codebase without a developer holding their hand at every step.

To find the answer, we ran a dual-track experiment following two PM/PD pairs on very different journeys: one starting a project from scratch and another attempting to build and ship products within a massive, pre-existing production environment filled with established patterns and strict architectural rules.

In this talk, we will pull back the curtain on the entire process. We will explain how we configured the initial setup to enable autonomy, the specific technical hurdles and logic-based problems we encountered when the AI hit its limits, and the crucial moments where the experiment failed and developers had to intervene to save the day. Finally, we will share how we assessed our confidence in the outcome&#8212;whether we truly reached a level of trust where the final code could be deployed&#8212;providing an honest look at whether the &quot;Product Engineer&quot; is a sustainable reality or a dangerous myth.</abstract>
                <slug>scbcn2026-226-testing-the-limits-can-pms-and-designers-build-products-in-the-real-world-</slug>
                <track>AI and Automation</track>
                
                <persons>
                    <person id='167'>David Valero</person><person id='171'>Rosana Reischak</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/C99XZP/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/C99XZP/feedback/</feedback_url>
            </event>
            <event guid='ea31014a-f0bd-5c19-bd8d-db440079f210' id='120'>
                <room>Track 1</room>
                <title>Secure by Design, Secure by Principles</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T17:30:00+02:00</date>
                <start>17:30</start>
                <duration>00:55</duration>
                <abstract>La inteligencia artificial no es solo una herramienta: es una infraestructura de poder. En esta charla exploramos c&#243;mo dise&#241;ar sistemas de IA que sean seguros desde su concepci&#243;n, no como parche posterior. Hablaremos de secure-by-design aplicado a IA, modelos robustos y predecibles, mitigaci&#243;n de riesgos (prompt injection, data poisoning, fugas de informaci&#243;n), gobernanza, soberan&#237;a tecnol&#243;gica y &#233;tica aplicada al dise&#241;o arquitect&#243;nico.

M&#225;s all&#225; de la ciberseguridad cl&#225;sica, abordaremos c&#243;mo integrar principios de ingenier&#237;a responsable, trazabilidad, explicabilidad y control humano significativo. El objetivo: construir sistemas confiables, auditables y alineados con valores humanos y sociales. Porque la seguridad no es una feature, es un principio de dise&#241;o, y de &#233;tica.</abstract>
                <slug>scbcn2026-120-secure-by-design-secure-by-principles</slug>
                <track>AI and Automation</track>
                
                <persons>
                    <person id='93'>Javier Ramos Panduro</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/LUX9W3/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/LUX9W3/feedback/</feedback_url>
            </event>
            
        </room>
        <room name='Track 2' guid='c2387d19-8add-5091-9ce1-391efd5121f8'>
            <event guid='e46b6a2c-7e8a-53a8-ba3e-e4cca6eb254c' id='206'>
                <room>Track 2</room>
                <title>The craft is still yours: rethinking junior growth with AI</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T11:30:00+02:00</date>
                <start>11:30</start>
                <duration>00:55</duration>
                <abstract>The narrative is spreading fast: AI is coming for the junior role. Why invest in growing someone when a model can generate a pull request in seconds?
This way of thinking is dangerous, and it&#8217;s costing us the next generation of great engineers. The real question was never &#8220;can AI do what a junior does?&#8221; but rather &#8220;how do juniors grow into seniors?&#8221; which is a question our industry has answered for decades with methods like pair programming.
With the right structure, AI becomes a natural evolution of that practice: a 24/7 companion that knows the project, answers without judgment, and meets the junior where they are. Spec-first and intentional, it keeps them in the driver&#8217;s seat while accelerating growth across multiple dimensions, from reasoning about requirements to product thinking to implementation and beyond.
Juniors aren&#8217;t just using AI; they&#8217;re managing it. They interrogate its output, catch its blind spots, and decide when to trust it or push back. That judgment is the craft, and it&#8217;s theirs to build.
This talk brings together two perspectives: leadership and lived experience. A junior AI engineer shares what it feels like, including what unlocked, what was hard, and what she wishes she&#8217;d known earlier. An industry leader adds perspective on designing the environment that makes this growth possible, and what breaks when structure is missing.
The craft is still yours, and AI is just another tool to grow it.</abstract>
                <slug>scbcn2026-206-the-craft-is-still-yours-rethinking-junior-growth-with-ai</slug>
                <track>AI and Automation</track>
                
                <persons>
                    <person id='23'>Diana Gamez</person><person id='153'>Lauren Tucker</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/MQNH7J/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/MQNH7J/feedback/</feedback_url>
            </event>
            <event guid='11216454-5343-5831-88f3-3d57fb7ed188' id='196'>
                <room>Track 2</room>
                <title>Building an Application Security function</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T12:30:00+02:00</date>
                <start>12:30</start>
                <duration>00:55</duration>
                <abstract>In a technology organisation of over 2,500 engineers across 200+ teams, building cloud-native, mobile, and embedded systems under external scrutiny (SOC 2, NIS2), we set out to create an Application Security function (sometimes named Product Security) that could scale without becoming a bottleneck, alongside an existing InfoSec capability.

This talk shares how we designed AppSec so that teams own the security of their applications, while a central team provides guidance, tooling, and enablement. It covers hiring for both technical skills and empathy, navigating funding challenges, and embedding security into delivery without slowing it down.

We will explore practical elements such as integrating dependency scanning into developer workflows; introducing threat modelling into standard development practices; and managing incidents, penetration testing, and bug bounty programmes. We also built a Security Champions community through training and hands-on initiatives.

Finally, we&apos;ll discuss the implications of the introduction of generative AI to the SDLC activities.

A central theme is culture: building trust with engineers and leadership, working collaboratively, and avoiding isolation. This experience offers insights for organisations looking to evolve Application Security as a shared, scalable responsibility in modern software development.</abstract>
                <slug>scbcn2026-196-building-an-application-security-function</slug>
                <track>Testing and Quality</track>
                
                <persons>
                    <person id='145'>Toni Tassani</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/Q9ZMXP/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/Q9ZMXP/feedback/</feedback_url>
            </event>
            <event guid='91806c81-c9d9-5506-9384-6ade9ae95267' id='155'>
                <room>Track 2</room>
                <title>Stop Designing Screens. Start Designing Decisions</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T15:00:00+02:00</date>
                <start>15:00</start>
                <duration>00:55</duration>
                <abstract>Modern software is no longer purely reactive.

From AI copilots to automated workflows, systems increasingly act on behalf of users&#8212;making suggestions, triggering actions, and sometimes making decisions without explicit input.

But most of our engineering practices still assume a simple model: request &#8594; response

So what happens when there is no clear request?

When the system decides to act, new challenges emerge:
-How do we model and implement decision-making in our systems?
-How do we observe and explain why something happened?
-How do we debug behavior that wasn&#8217;t directly triggered by a user?
-How do we define requirements for actions that may or may not happen?

In this talk, we&#8217;ll explore how agentic behavior changes the way we design, build, and operate software.

We&#8217;ll look at:
-how to move from endpoint-driven thinking to decision-driven architecture
-what &#8220;observability&#8221; means when systems make choices&#8212;not just process inputs
-practical ways to keep autonomous behavior traceable, testable, and safe

This is not a UX talk. It&#8217;s a shift in how we think about systems. Because if your software can decide to act&#8212;and you didn&#8217;t design how&#8212;you&#8217;ll still get decisions. Just ones you can&#8217;t see, explain, or control.</abstract>
                <slug>scbcn2026-155-stop-designing-screens-start-designing-decisions</slug>
                <track>Monitoring and observability</track>
                
                <persons>
                    <person id='71'>Nastassia Shahun</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/3P7AZW/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/3P7AZW/feedback/</feedback_url>
            </event>
            <event guid='60706dce-5535-560c-880a-4d7f77f3adb2' id='219'>
                <room>Track 2</room>
                <title>Exactly-Once Semantics: Getting It Right When Money Is on the Line</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T16:00:00+02:00</date>
                <start>16:00</start>
                <duration>00:55</duration>
                <abstract>Exactly-once semantics is a cornerstone - and often unspoken - rule of many business requirements: we must not charge a user twice for the same product, deliver the same service twice, or lose a payment halfway through.

But the distributed systems we build on rarely provide exactly-once behavior end to end. As business rules grow more complex and architectures span services, queues, databases, and third-party APIs, getting this right becomes a really hard engineering problem.

In this talk, we&#8217;ll walk through a real fintech case from the initial product promise to the technical architecture behind it. We&#8217;ll bridge the gap between business expectations and the guarantees engineers must build, examine the most important failure modes, and arrive at a solution that delivers effectively-once behavior in practice.</abstract>
                <slug>scbcn2026-219-exactly-once-semantics-getting-it-right-when-money-is-on-the-line</slug>
                <track>Software Design and Architecture</track>
                
                <persons>
                    <person id='163'>Sergei Aleksandrov</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/M8XKK3/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/M8XKK3/feedback/</feedback_url>
            </event>
            <event guid='c5598452-0fa0-5f9a-87c5-cd738b990b1f' id='137'>
                <room>Track 2</room>
                <title>My journey as a developreneur: accessible games and apps that work for everyone</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T17:30:00+02:00</date>
                <start>17:30</start>
                <duration>00:55</duration>
                <abstract>Hello.
I started as a developer in 2020 after many years of wanting to make mobile games and apps for everyone (I&apos;m a solo totally blind developer).

This talk is not about becoming or a developer or about developing as a blind person. Rather, it is a talk to share my experience, give advice to new devs, and showcase what I&apos;ve tried to do over the years. Failures, successes, future projects, and how AI is changing our lives as developers and, yes, how AI is helping blind developers make better UIs.</abstract>
                <slug>scbcn2026-137-my-journey-as-a-developreneur-accessible-games-and-apps-that-work-for-everyone</slug>
                <track>Diversity in IT</track>
                
                <persons>
                    <person id='111'>Oriol G&#243;mez Sent&#237;s</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/JPMZRQ/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/JPMZRQ/feedback/</feedback_url>
            </event>
            
        </room>
        <room name='Track 3' guid='7b4a0a1e-2868-5eee-bd98-bdc42bef3917'>
            <event guid='b0244766-493c-5811-a3e5-17d31c526918' id='208'>
                <room>Track 3</room>
                <title>Tu pr&#243;ximo pair puede no saber programar</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T11:30:00+02:00</date>
                <start>11:30</start>
                <duration>00:55</duration>
                <abstract>Durante d&#233;cadas, la frontera entre &quot;quienes construyen&quot; y &quot;quienes deciden qu&#233; construir&quot; estaba clara. Un PM escrib&#237;a un PID, ingenier&#237;a estimaba, y meses despu&#233;s sal&#237;a algo a producci&#243;n. La IA ha dinamitado esa frontera.
En Cabify lo estamos viviendo en primera persona. Product Managers que prototipan soluciones funcionales en horas. Designers que validan ideas con datos sint&#233;ticos antes de que ingenier&#237;a escriba una l&#237;nea de c&#243;digo. Sesiones de pair programming donde un IC y un perfil de producto trabajan juntos en Cursor, combinando visi&#243;n de negocio con criterio t&#233;cnico.
Pero este escenario no es gratis. Sin guardrails claros, el resultado es caos disfrazado de velocidad. El rol de ingenier&#237;a no desaparece: muta. Pasamos de escribir todo el c&#243;digo a dise&#241;ar las reglas del juego: coberturas de test, est&#225;ndares de calidad, pipelines de validaci&#243;n y l&#237;mites claros sobre qu&#233; llega a producci&#243;n y qu&#233; no.
Compartir&#233; historias reales de c&#243;mo estamos redefiniendo nuestro flujo: qu&#233; ha funcionado, qu&#233; ha explotado, qu&#233; guardrails hemos construido tras problemas que no anticipamos, y c&#243;mo institucionalizamos una forma de trabajar donde prototipar es casi gratis pero la calidad no se negocia. No es una charla sobre herramientas. Es sobre qu&#233; pasa cuando toda la organizaci&#243;n puede construir producto, y qu&#233; debe hacer ingenier&#237;a para que esa velocidad no se convierta en caos.</abstract>
                <slug>scbcn2026-208-tu-prximo-pair-puede-no-saber-programar</slug>
                <track>Innovation and Trends</track>
                
                <persons>
                    <person id='39'>Diego Albert</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/THQCZY/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/THQCZY/feedback/</feedback_url>
            </event>
            <event guid='be8fa494-7aa7-59e6-9d0b-7c8836095952' id='174'>
                <room>Track 3</room>
                <title>Cuando Espa&#241;a se apag&#243;: arquitectura para sobrevivir cuando falla hasta la luz</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T12:30:00+02:00</date>
                <start>12:30</start>
                <duration>00:55</duration>
                <abstract>Dise&#241;amos sistemas asumiendo que la conectividad es un hecho. El a&#241;o pasado, durante el apag&#243;n nacional, esa suposici&#243;n se rompi&#243;. Algunas partes de nuestros sistemas aguantaron. Otras no. Y la diferencia entre unas y otras no fue suerte: fueron decisiones de arquitectura tomadas a&#241;os antes, en fr&#237;o, cuando nadie quer&#237;a justificar el coste y no hab&#237;a forma de probar que val&#237;an la pena.

En esta charla cuento esas decisiones, con los trade-offs de negocio detr&#225;s de ellas, que permitieron que nuestros almacenes siguieran operando sin conexi&#243;n a sistemas centrales durante horas. Con 1.700 tiendas, producto perecedero y log&#237;stica que cuesta miles por minuto de parada.

Tres problemas concretos: el modo bunker (c&#243;mo un nodo opera solo), el split-brain (qu&#233; pasa cuando la realidad f&#237;sica y la base de datos central divergen durante horas, y qui&#233;n gana) y la reconciliaci&#243;n (c&#243;mo resincronizar millones de transacciones oscuras sin p&#233;rdida de datos cuando el sistema vuelve a la vida, y los race conditions que tienes que haber dise&#241;ado antes de que ocurran).

Cada abstracci&#243;n t&#233;cnica de esta historia tiene una restricci&#243;n de negocio detr&#225;s. La charla no separa las dos cosas.</abstract>
                <slug>scbcn2026-174-cuando-espaa-se-apag-arquitectura-para-sobrevivir-cuando-falla-hasta-la-luz</slug>
                <track>Software Design and Architecture</track>
                
                <persons>
                    <person id='43'>Emilio Carri&#243;n</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/RWYFFM/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/RWYFFM/feedback/</feedback_url>
            </event>
            <event guid='42324132-f691-5132-9903-3d840732cd78' id='139'>
                <room>Track 3</room>
                <title>Kit de supervivencia para fechas l&#237;mite: Optimizando las rutas de 22.000 pedidos bajo presi&#243;n</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T15:00:00+02:00</date>
                <start>15:00</start>
                <duration>00:55</duration>
                <abstract>A todos nos ha pasado eso de llegar un d&#237;a a la oficina, y antes siquiera de poder tomarte un caf&#233; alguien te avisa de que hay una reuni&#243;n urgente. Te acercas temoroso, pues sabes que se avecinan curvas. 

En nuestro caso el negocio ha crecido, el motor de rutas que fue construido al inicio del proyecto ya no soporta los 22.000 pedidos que debe entregar a diario. Crea rutas ineficientes que requieren de manualidades provocando retrasos en las entregas, clientes descontentos y un equipo exhausto haciendo horas extra. 

Se trata de una situaci&#243;n peliaguda, y los stakeholders lo necesitan &quot;para ayer&quot;. &#191;Qu&#233; haces ante esta situaci&#243;n?

En esta charla abriremos las tripas de un caso real donde la **mentalidad Lean** y las pr&#225;cticas de **Extreme Programming (XP)** fueron nuestro salvavidas. Veremos c&#243;digo, m&#233;tricas y gr&#225;ficas para detallar todo el camino que seguimos.

&#191;Qu&#233; veremos en la sesi&#243;n?

- Analizaremos las opciones que ten&#237;amos y por qu&#233; elegimos nuestra soluci&#243;n actual, siempre abogando por una soluci&#243;n pragm&#225;tica con sus tradeoffs.
- C&#243;mo gestionamos las expectativas de los stakeholders.
- C&#243;mo las din&#225;micas de equipo (feedback constante, simplicidad y coraje) nos permitieron atacar el problema sin quemarnos.
- C&#243;mo utilizar la &#8220;deuda t&#233;cnica&#8221; como herramienta de entrega eficiente sin que fuera un obst&#225;culo.
- Qu&#233; pas&#243; en la primera puesta en producci&#243;n y c&#243;mo nuestras decisiones iniciales afectaron al sistema meses despu&#233;s.
- Un repaso por lo que sali&#243; bien y, sobre todo</abstract>
                <slug>scbcn2026-139-kit-de-supervivencia-para-fechas-lmite-optimizando-las-rutas-de-22-000-pedidos-bajo-presin</slug>
                <track>Development principles and methodologies</track>
                
                <persons>
                    <person id='113'>Alejandro</person><person id='115'>Victor Nassif</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/RV9H7F/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/RV9H7F/feedback/</feedback_url>
            </event>
            <event guid='39c4fa5e-cd11-5b55-b260-b2122d219d79' id='178'>
                <room>Track 3</room>
                <title>De vibe coding a harness engineering: modernizando un entorno regulado con la IA</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T16:00:00+02:00</date>
                <start>16:00</start>
                <duration>00:55</duration>
                <abstract>Migrar un mainframe bancario a la nube bajo normativas de seguridad fluctuantes en tres meses parec&#237;a una misi&#243;n imposible. La IA (agentes, herramientas MCP) fue el multiplicador de fuerza para lograrlo, pero exigi&#243; un cambio de paradigma. Esta sesi&#243;n disecciona la evoluci&#243;n desde la ilusi&#243;n del vibe coding &#8212;generar c&#243;digo mediante conversaciones fluidas&#8212; hacia una estricta disciplina de harness engineering (ingenier&#237;a de arneses).

Cuando la IA escribe el c&#243;digo, la artesan&#237;a del software (Software Craftsmanship) evoluciona. El foco ya no es la sintaxis, sino la construcci&#243;n de las redes de seguridad. Exploraremos c&#243;mo domar la generaci&#243;n autom&#225;tica en entornos hostiles y altamente regulados construyendo arneses implacables: Spec-Driven Development, pipelines evolutivos y validaci&#243;n continua.

Puntos Clave
Evoluci&#243;n del Craftsmanship: El paso de teclear c&#243;digo a orquestar, limitar y auditar sistemas generados por m&#250;ltiples subagentes.

Construcci&#243;n de Arneses: Uso de testing extremo (integraci&#243;n, testcontainers, simulaci&#243;n de entornos) para contener y certificar el comportamiento del c&#243;digo generado.

Spec-Driven Development: La especificaci&#243;n como contrato &#250;nico e inquebrantable entre negocio, seguridad y los agentes de IA.

Pragmatismo radical: C&#243;mo mantener la agilidad y los principios de dise&#241;o cuando las reglas del entorno regulado cambian a diario.</abstract>
                <slug>scbcn2026-178-de-vibe-coding-a-harness-engineering-modernizando-un-entorno-regulado-con-la-ia</slug>
                <track>AI and Automation</track>
                
                <persons>
                    <person id='7'>Anyul Rivas</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/UF8SFR/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/UF8SFR/feedback/</feedback_url>
            </event>
            
        </room>
        <room name='Track 4' guid='3aa5f09a-28bb-59e6-b855-135a31473c5b'>
            <event guid='14fc0530-b2ef-5e2d-a888-2d973b3789af' id='223'>
                <room>Track 4</room>
                <title>Documentando APIs para IAs</title>
                <subtitle></subtitle>
                <type>Lighning talk</type>
                <date>2026-10-17T11:30:00+02:00</date>
                <start>11:30</start>
                <duration>00:07</duration>
                <abstract>El a&#241;o pasado en la SCBCN25 habl&#233; de c&#243;mo Documentar APIs. Dej&#233; fuera una parte que estaba incipiente y ha explotado en doce meses: ahora quien integra tu API muchas veces no es una persona, es una IA.

En esta lightning quiero hacer un repaso r&#225;pido del panorama actual:
&#10143; Herramientas y est&#225;ndares que ampl&#237;an el ecosistema: MCPs y generaci&#243;n de tool definitions, llms.txt, OpenAPI para agentes
&#10143; Casos de uso: chatbots con RAG sobre la documentaci&#243;n, agentes de soporte&#8230; 
&#10143; Qu&#233; cambia en c&#243;mo documentamos

Y por supuesto, cualquier novedad rese&#241;able, que las habr&#225; desde la presentaci&#243;n de este C4P hasta el d&#237;a del evento &#128521;</abstract>
                <slug>scbcn2026-223-documentando-apis-para-ias</slug>
                <track>AI and Automation</track>
                
                <persons>
                    <person id='83'>Javi Rubio</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/LKJLFB/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/LKJLFB/feedback/</feedback_url>
            </event>
            <event guid='03a502f9-c9c3-563f-8605-dd5eca992c58' id='152'>
                <room>Track 4</room>
                <title>Rescuing the System Everyone Gave Up On</title>
                <subtitle></subtitle>
                <type>Lighning talk</type>
                <date>2026-10-17T11:44:00+02:00</date>
                <start>11:44</start>
                <duration>00:07</duration>
                <abstract>What happens when only 3 out of 100 engineers can touch a critical system and nobody else wants to? This talk dives into a real story of a failing legacy codebase: programming language barrier, poor testing, outdated dependencies, constant outages and a team stuck in frustration with no investment in sight.
Instead of another failed &#8220;big rewrite&#8221; proposal, I approached the problem differently, using product thinking. By breaking the system into smaller, prioritized problems, aligning stakeholders and introducing measurable outcomes, we turned paralysis into progress.
In this lightning talk, I&#8217;ll walk through the exact steps I used to reframe the problem, gain buy-in and start delivering impact without having to think about rewriting everything. If you&#8217;ve ever faced a legacy system everyone avoids but depends on, this talk will give you a practical way forward.</abstract>
                <slug>scbcn2026-152-rescuing-the-system-everyone-gave-up-on</slug>
                <track>Tech culture and soft skills</track>
                
                <persons>
                    <person id='121'>Andrei Danilov</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/SEEMUQ/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/SEEMUQ/feedback/</feedback_url>
            </event>
            <event guid='f7eb1723-7349-5057-a255-514050bd915c' id='205'>
                <room>Track 4</room>
                <title>What PLG Actually Means for an Engineering Team</title>
                <subtitle></subtitle>
                <type>Lighning talk</type>
                <date>2026-10-17T12:30:00+02:00</date>
                <start>12:30</start>
                <duration>00:07</duration>
                <abstract>You won&apos;t have a thorough PRD and months to implement a feature. You have a hypothesis to validate, a quick solution to ship, and a few days later you&apos;re measuring its impact. This talk covers what PLG engineering actually looks like in practice: from running quick experiments and negotiating scope, to measuring what matters and avoiding the dead code traps of A/B testing.</abstract>
                <slug>scbcn2026-205-what-plg-actually-means-for-an-engineering-team</slug>
                <track>Tech culture and soft skills</track>
                
                <persons>
                    <person id='152'>Marta Vasconcelos</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/GUH8VX/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/GUH8VX/feedback/</feedback_url>
            </event>
            <event guid='b676e7a9-8aea-53f2-91ec-abba045c14a8' id='222'>
                <room>Track 4</room>
                <title>La feature perp&#232;tua: Com integrar accessibilitat i complir l&apos;EAA amb l&apos;ajuda de la IA</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T16:00:00+02:00</date>
                <start>16:00</start>
                <duration>00:55</duration>
                <abstract>L&#8217;accessibilitat &#233;s la &quot;feature perp&#232;tua&quot;: aquell tiquet que viu eternament al backlog, s&#8217;hereda de sprint en sprint i mai no es tanca perqu&#232; &quot;no hi ha temps&quot;. Per&#242; el 2026, amb l&#8217;EAA (European Accessibility Act) en ple vigor, ajornar-la ja no &#233;s una opci&#243; t&#232;cnica, &#233;s un risc legal imminent. La realitat &#233;s que la majoria dels equips no estan preparats i confien en automatitzacions b&#224;siques que nom&#233;s generen una perillosa il&#183;lusi&#243; de qualitat.
&#8203;En aquesta sessi&#243;, trencarem el cicle de postergaci&#243;. Analitzarem per qu&#232; els validadors tradicionals fallen en detectar barreres reals i com l&#8217;Enginyeria d&#8217;Agents revoluciona el compliment de la normativa. Explorarem com l&#8217;orquestraci&#243; d&#8217;agents d&apos;IA personalitzats, creats espec&#237;ficament per a auditar l&apos;accessibilitat, permet raonar sobre la sem&#224;ntica i la interacci&#243; humana. Mitjan&#231;ant el Model Context Protocol (MCP) i les Chrome DevTools, veurem com aquests agents auditen amb precisi&#243; all&#242; que els tests autom&#224;tics ignoren.
&#8203;Descobrirem una estrat&#232;gia per integrar auditors intel&#183;ligents que eleven la capacitat dels equips, transformant l&#8217;accessibilitat d&#8217;una c&#224;rrega inabastable en un proc&#233;s t&#232;cnic escalable, auditable i real. &#201;s hora de deixar de moure el tiquet de sprint i comen&#231;ar a garantir una web que no exclogui ning&#250;.</abstract>
                <slug>scbcn2026-222-la-feature-perptua-com-integrar-accessibilitat-i-complir-l-eaa-amb-l-ajuda-de-la-ia</slug>
                <track>Accesibility and sustainability</track>
                
                <persons>
                    <person id='164'>Roc&#237;o Cejudo Soronellas</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/U7MKSM/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/U7MKSM/feedback/</feedback_url>
            </event>
            <event guid='73b306cf-7d4c-5f2d-b66c-ef430d3d445f' id='132'>
                <room>Track 4</room>
                <title>Asteroide A11y-B-612</title>
                <subtitle></subtitle>
                <type>Talk</type>
                <date>2026-10-17T17:30:00+02:00</date>
                <start>17:30</start>
                <duration>00:55</duration>
                <abstract>Cada any, WebAIM estudia l&#8217;estat global de l&#8217;accessibilitat analitzant un mili&#243; de p&#224;gines web. En l&#8217;an&#224;lisi de l&#8217;&#250;ltim any (publicada al febrer), es van detectar prop de 56 milions d&#8217;errors diferents. El m&#233;s curi&#243;s &#233;s que el 96 % d&#8217;aquests errors es poden agrupar en sis categories, que han estat les mateixes durant els darrers set anys pero l&apos;&#250;ltim any han empitjorat. Vols saber qu&#232; ha passat i quines s&#243;n?

Aquesta xerrada et servir&#224; per entendre com afecta l&#8217;accessibilitat als teus projectes digitals i com evitar els errors m&#233;s habituals. Al llarg del viatge, visitarem diversos planetes d&#8217;El Petit Pr&#237;ncep, on els seus personatges ens ajudaran a descobrir conceptes clau d&#8217;accessibilitat, bones pr&#224;ctiques i algunes lli&#231;ons imprescindibles per construir experi&#232;ncies digitals m&#233;s inclusives.</abstract>
                <slug>scbcn2026-132-asteroide-a11y-b-612</slug>
                <track>Accesibility and sustainability</track>
                
                <persons>
                    <person id='106'>Jordi Turull</person>
                </persons>
                <language>en</language>
                
                <recording>
                    <license></license>
                    <optout>false</optout>
                </recording>
                <links></links>
                <attachments></attachments>

                <url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/TMFS7J/</url>
                <feedback_url>https://c4p.softwarecrafters.barcelona/scbcn2026/talk/TMFS7J/feedback/</feedback_url>
            </event>
            
        </room>
        
    </day>
    
</schedule>
