Team
Tuxfram
Project Concept
No description has been added yet.
Entry
Status: Submitted
Last saved: May 09 at 5:46 PM -05
Team Roster
You must be registered for the event to view the team message board.
Juan José Duque Londoño Team Lead RSVP Approved
Analisis y desarrollo de software at SENA
Responsible for the frontend UI design and the session progress experience. Built the MeditationGuide.tsx component including the animated breathing visualizations, countdown timer, streak tracker, and post-session feedback panel. Implemented the glassy pastel design system using Tailwind CSS and ensured the interface responded fluidly to each agent tool call. Also handled the overall Next.js project structure, routing, and deployment configuration.
Soy un estudiante apasionado por la tecnología; me encanta aprender cosas nuevas cada día. Me considero un gran entusiasta de este campo y estoy en búsqueda de experiencias que contribuyan a mi desarrollo profesional.
Actualmente, mi mayor interés se centra en el ecosistema de la inteligencia artificial y en cómo puede aplicarse a mi área favorita: la ciberseguridad.
Actualmente, quiero dedicarme a un proyecto de creación de una distribución de Linux basada en Arch, enfocada en la ciberseguridad. Esta distribución tendría como característica principal la integración de inteligencia artificial directamente en el sistema operativo, con el objetivo de guiar y apoyar a estudiantes de ciberseguridad durante su aprendizaje.
Juan Pablo Egido Castellanos RSVP Approved
Product Engineer at Lumai
Led the overall architecture and technical direction of the project. Designed and implemented the Agent-to-UI (A2UI) tool-calling system in src/lib/toolDeclarations.ts, defining all four AI tool declarations (recommend_duration, set_meditation_duration, begin_meditation_timer, end_meditation_feedback) that allow the Gemini agent to trigger real-time UI state changes. Coordinated the integration between the AI layer and the React component tree, ensuring the agent's outputs translated directly into functional, dynamic interface updates.
Juan Pablo Egido Castellanos is the Co-Founder of Stealth Startup and a Product Engineer at Lumai, where he is self-employed and involved in Tinkerer projects like lumai.space as an AI Engineer focusing on Application Development. He has approximately 11 years of experience and studied Ingeniería informática at Universidad de Castilla-La Mancha and Business/Management at EOI Business School. Based in Medellín, Antioquia, Colombia, Juan Pablo is looking for customers, co-founders (technical and business), investors, and speaking opportunities, and is open to intros via DM. He can also be found on Twitter as @juanegido.
Partnerships, product review, technical architecture, technical co-founders, business co-founders, investors, AI/ML systems, conversational AI, data engineering, AI-enabled products, backend microservices, application development.
lumai.space
Angel Mateus RSVP Approved
Analista at the school of breath
Responsible for integrating the Google Gemini Live API's real-time voice and audio capabilities into the application. Built the live voice interaction pipeline that allows the AI agent to listen to the user through the microphone and respond with guided audio during meditation sessions. Implemented the WebRTC and MediaStream API setup for microphone, camera, and screen share controls, and designed the ControlTray component with its glassmorphism circular UI using Feather Icons.
I'm a web developer with a Computer Science degree and 6+ years of experience crafting interactive and accessible applications. Skilled in JavaScript frameworks like React, Next.js, Vue, and Nuxt, I bring intuitive designs to life with HTML, CSS, and SVG. I'm passionate about language exchange and am the creator of eventsmedellin community. Let's create something remarkable!
networking , work , share projects
i'm doing chat bot ai