Lead Mobile & AI Engineerfor Founders Who Need to Ship
Not a founder. The engineer founders call when the technical calls get hard — native vs hybrid, AI runtime vs monolith, ship fast vs ship right.
Independent product engineer working with founders and studios. Currently shipping with CODX — ChutApp, Kunoa, S-Core. Based in Tenerife, between Bangkok and Shenzhen.
Jiménez
Three with CODX, one on the road.
Mobile-first products shipped between 2023–2025 with CODX's studio practice, plus a solo experiment from Asia. Each one made a different call — native, hybrid, AI-driven, or living inside a closed Chinese ecosystem. The calls are the work.
- App / 01CollabLive
ChutApp
Social video platform for youth football
RoleLead mobile engineerPlayers, teams and families share clips, goals and training moments in a moderated environment built for under-18s. The surface is video, the feel is native — gesture timing, capture quality, playback consistency.
Key callInsisted on dual-platform native from day one instead of hybrid shortcuts. Short-form video lives on device feel — gesture timing, capture quality, playback consistency. We chose to ship every feature twice and keep that surface tight.
Impact- 12,000 users on the platform
- 40% monthly retention
- Content moderation system I designed
SwiftKotlinNext.jsAWS AmplifyBuilt at CODX · codx.digital↗Year2023 — NowRoleLead mobile engineerOutputiOS + Android, live - App / 02CollabLive
Kunoa
AI coaching with expert avatars
RoleBuilt the AI personalisation layerTrainers and nutritionists become AI avatars the user interacts with. The plan adapts; the voice doesn't. I built the personalisation layer — the models that learn, the runtime that delivers, the seam that lets them evolve.
Key callSplit the AI runtime from the product runtime. FastAPI carries the personalisation models; Next.js carries the marketing and admin surface. The seam lets the model evolve weekly without app redeploy.
Impact- 35% churn reduction with AI personalisation
- Trainers onboarded in 2 days
SwiftKotlinFastAPIAWSPostgreSQLBuilt at CODX · codx.digital↗Year2024RoleAI + mobile integrationOutputiOS + Android, live - App / 03CollabLive
S-Core
Gamified fan communities for DJs
RoleLed payments and platform decisionsFans complete pre-event challenges to unlock tickets, drops and exclusive access. Multi-currency rewards, event-aware feeds. I led the payments layer and made the call that shaped the Android surface.
Key callReact Native on Android instead of native Kotlin. The surface is content and transactions, not gesture-heavy — and four continents asked for feature parity faster than dual-platform native could deliver. Different product, different call.
Impact- Live across 4 continents
- Zero payment failures at scale
SwiftReact NativeNode.jsFirebaseStripeBuilt at CODX · codx.digital↗Year2025RolePayments + platformOutputLive across 4 continents - App / 04SoloBuilt · Unpublished
China Survival Guide
WeChat Mini Program for expats in China
RoleSolo buildA guide for foreigners moving through China, shipped inside WeChat as a Mini Program. Two views, single content store, no backend. Built and tested end-to-end inside WeChat's runtime.
Key callStayed inside WeChat's own primitives — WXML, WXSS, JavaScript. No bridge layer, no React shim, no backend. The point was to learn how the system actually works from the inside, not to bend it back into something familiar.
Impact- Working build validated with 50+ expats
- Publication blocked by entity requirement only
WXMLWXSSJavaScriptWeixin DevtoolsRead the build log↗Year2026RoleSolo buildOutputWorking build · Unpublished
Not a founder. The engineer you call when the calls get hard.
Three principles that shape how I work with founders and studios. Execution over ownership. Decisions over decks. Results over resumes.
I don't own these products. I own the decisions.
At CODX, I was the lead mobile and AI engineer — the person founders talked to when they needed to decide: native or hybrid? Separate AI runtime or monolith? These calls are the work. The products are the proof.
Collaboration, not foundership.
The founders had the vision. I had the technical execution. ChutApp needed video that felt native — I shipped dual-platform from day one. Kunoa needed AI that evolved weekly — I split the model runtime from the app. That's what I do.
Results I delivered.
- 12,000 users on a social video platform
- 35% churn reduction with AI personalisation
- 4-continent launch in 8 weeks
Currently building.
Personal vehicles in different states of becoming. Not portfolio pieces — ongoing work the studio practice keeps coming back to.
- Building / 01Waitlist live
QuestClub
Private social game for friend groups built around seasonal missions, shared memories and long-term group competition.
StatusOnboarding systems and gameplay loops in active development.
questclub.app↗ - Building / 02Daily use
LanguageOS
Personal language-learning system focused on immersion, cultural context and daily usage rather than traditional gamification.
StatusInternal tool in active daily use.
language-os.vercel.app↗ - Building / 03In formation
Tian Core
Creative systems studio exploring AI-native products, editorial interfaces and internet culture through a cinematic digital aesthetic.
StatusIdentity system + experimental web explorations.
tian-core.vercel.app↗
Retainer-first, not project-based.
Ongoing technical leadership for mobile and AI products. Fixed scope sprints available to test the fit. Everything starts with a 15-minute call.
- Engage / 01
Fractional Mobile + AI Engineer
Ongoing·$2,000 per month·3-month minimum, then month-to-monthYour lead engineer on retainer. I maintain, evolve, and add AI features to your mobile product.
For:Startups with an app in production that need technical leadership without hiring full-time.
Book a 15-min call↗Scope- 15–20 hours per week
- Bug fixes and performance optimisation
- 1 new AI feature per month
- Architecture decisions and code review
- Direct Slack access, 24h response time
Deliverables- Weekly progress updates
- Monthly roadmap review
- Code handover documented
- Engage / 02
AI Integration Sprint
2 weeks·$2,500 one-time·Fixed scope, fixed timelineOne AI workflow mapped, one production-ready integration shipped. Perfect for testing the relationship before a retainer.
For:Teams adding AI to an existing product without losing a quarter to it.
Start a sprint↗Scope- Workflow audit and integration spec
- One LLM, agent, or automation in production
- Eval harness for quality measurement
- Handover doc for your team
Deliverables- Working integration
- Documentation
- 30-day support
Not sure which one? Let's talk.
Start a conversation→Working notes on mobile, AI & Asia.
Personal voice — drafted in Spanish on purpose. Mobile dev decisions, agentic-AI commentary, and what I find moving through Asian tech ecosystems. English work and client copy lives everywhere else on the site.
- 2026 · 07 · 13Field log
Susurro: construí control por voz 100% local para Claude Code
Tenía diez sesiones de Claude Code abiertas a la vez y no daba abasto mirando pantallas. Construí un sistema de voz que dicta, avisa y cambia de sesión — transcripción y voz corriendo 100% local en el Mac, sin API keys.
Read → - 2026 · 04 · 28Field log
Mi experiencia creando una mini app en WeChat
Una guía de China para extranjeros, construida dentro de WeChat. Lo complicado no fue programar — fue entender el sistema cerrado.
Read → - 2026 · 04 · 04Essay
Gemma 4: Google libera su modelo mas potente bajo Apache 2.0
Google lanza Gemma 4 bajo licencia Apache 2.0 por primera vez, con modelos edge que corren en moviles y Raspberry Pi. Analisis tecnico y estrategico.
Read → - 2026 · 04 · 01Essay
IA Agentica en 2026: El Fin de las Apps Tal Como las Conocemos
Los agentes de IA estan redefiniendo la interaccion con el software. El 40% de las empresas tendran agentes operativos en 2026, y el Model Context Protocol esta creando un nuevo estandar. Que significa esto para los desarrolladores.
Read → - 2026 · 03 · 31Essay
La IA ya no solo escribe codigo: ahora tambien lo audita
Qodo cierra una ronda de 70 millones para verificar codigo generado por IA. El desarrollo de software se divide en dos capas: generacion y verificacion.
Read →