Georges S.

Design and deliver the platform's front-end, from the design system to production deployment on the defined domain, integrating it with the cryptographic verification back-end.

Created a complete design system (tokens, components, states) and developed 6 responsive pages for the public interface and the integrator console. Integrated with the back-end via a typed TypeScript client layer covering all verification states, from nominal cases to error scenarios (expiration, source unavailability, integrity failure). Connected the two key workflows: public proof verification by hash and administration console for integrators (API key, history). Deployed to the final domain via Vercel, with security headers, end-to-end tests, and handoff documentation for maintenance.

HTML, CSS, TypeScript, Vercel, REST, ECDSA, JWKS.

Design and deliver the back-end for a cryptographic proof verification platform, enabling third parties to independently authenticate data from external sources (insurers, aggregators, certification bodies).

Set up a strict TypeScript architecture with a secure REST API (Zod validation, API key/JWT authentication) and a PostgreSQL database for verifications and proofs. Developed an adaptable HTTP connector layer for any external source, with documented interface contracts. Exposed a public endpoint for verification by hash and ECDSA signature, allowing third parties to authenticate a proof without an API key. Implemented a full CI/CD pipeline, integration tests, and OpenAPI/Swagger documentation. Delivered a production-ready API, capable of receiving real data without engine modifications.

TypeScript, Node.js, PostgreSQL, Zod, JWT, ECDSA, OpenAPI/Swagger, CI/CD.

Design and deliver an AI voice assistant SaaS platform capable of managing phone calls, SMS, emails, and business automations end-to-end.

Set up the complete architecture (Vite/React front-end, Node.js/Express back-end, PostgreSQL database with Prisma), secure JWT authentication, Docker deployment on Hetzner with nginx and SSL. Integrated Twilio Programmable Voice and a direct voice pipeline Azure STT → GPT-4o-mini → Azure TTS, dynamic prompts, call history, and AI summaries. Connected Stripe Billing (subscriptions, trials, webhooks) and secured endpoints. Developed the user dashboard (statistics, history, number management, AI settings), additional modules (WhatsApp Business, Gmail/Outlook, PDF generation, multilingual, Zapier/Make/n8n connectors), Bull/Redis workers for campaigns and automations, Sentry/Pino monitoring, and complete technical documentation. Delivered a stable V1 in production.

React, TypeScript, Node.js, Express, PostgreSQL, Prisma, Twilio, Azure Speech, OpenAI, Stripe, Docker, Hetzner, nginx, Redis, Bull, n8n.