Jc Alhinho

Design and development of Elysée Arena 2027, a real-time satirical-political simulation allowing visualization of how 8 archetypes of candidates would react to any news (text or URL), with a "live" interface rendering. Implementation of the full-stack architecture: Python/FastAPI backend exposing a Server-Sent Events (SSE) stream to broadcast analysis and reactions in real-time, orchestration of an AI pipeline (analyst + candidates) on Google Gemini (text generation + illustration), and integration of a RAG engine based on ChromaDB (programs segmented into chunks, filtering by candidate, embeddings) to ground responses in a knowledge base. Containerized deployment on Google Cloud Run with GitHub Actions CI/CD (build/push to Artifact Registry, variable injection, CORS), storage of assets and results (illustrations, votes) on Cloud Storage, and "production-grade" security (rate limiting, input validation/cleaning, CSRF token, basic prompt-injection detection). Frontend: React + TypeScript (Vite), mobile-first UI with animations (Framer Motion), Markdown rendering, grid/carousel view, "knowledge base" modals, and anonymous voting system via browser fingerprint (Canvas/WebGL + SHA-256).

Tech: React, Material UI, Framer Motion

Design and development of a web application to modify one or more photos in 3 simple steps using AI (artistic styles, group photos, visual identity). Implementation of the Node/Express backend: REST routes, Gemini API integration, credit management, AI rendering logic, and guardrails. React/TypeScript frontend: guided UX, configurable presets, real-time integration of AI responses, Framer Motion animations. Hosting on GCP (Cloud Run / Functions, Firestore) with GitHub Actions CI/CD pipeline for automatic build, testing, and deployment.

Tech: React, Tailwind, ECharts, Grafana, Superset, Rust APIs

Development of real-time network supervision dashboards for SOC centers, with strong constraints on readability and performance. Integration of interactive ECharts graphs (timeseries, multi-layer scatterplots) supporting over 5,000 data points without loss of fluidity. Connection to internal Rust APIs: rights management, advanced filters, pagination, JSON payload optimization. Implementation of a design system documented in Storybook, modular and typed code, testing with Vitest. Contribution to CI/CD (GitLab CI), optimization of initial loading time and high-volume rendering.