Summary
Full-stack engineer with 9+ years shipping features end to end, from data model and API to UI. Senior in frontend (React, Next.js, TypeScript), mid-level in backend (Node.js, NestJS, PostgreSQL).
I built Mentoma alone, a multi-tenant SaaS with tenant isolation enforced in the database, RBAC, idempotent payments and a 14-day trial into paid plans. At Mcinext I delivered real-time analytics features, a 90% React performance gain and a testing culture from 0 to 80% coverage.
I build with coding agents (Claude Code, Codex, Cursor) every day, with human review on every change. Today I work as an AI Technical Product Manager, taking LLM, RAG and agent ideas from discovery and Python MVPs to delivery.
Experiences
AI Technical Product Manager
2026
- present
- Collaborate with AI, backend and frontend teams on architecture trade-offs and end-to-end delivery
- Work on the AI team, whose AI products and open-weight models run in production
- Own AI product development from technical discovery and Python MVP prototyping through validation and delivery, on a React/Next.js and Python stack
- Evaluate LLM, RAG and agent-based approaches and validate technical feasibility before productization
- Translate business requirements into technical AI specs, MVP scope and non-functional requirements for engineering
- LLM
- RAG
- Agents
- Python
- Product discovery
- Next.js
Senior Frontend Engineer → Agentic Frontend Engineer
October 2021
- 2026
- Designed and delivered full-stack features end-to-end (database → API → UI) for AI-powered real-time analytics platforms processing millions of daily events using React, TypeScript, Next.js and Node.js
- Worked as an agentic frontend engineer for the past two years: built, refactored and tested features with coding agents (Claude Code, Codex, Cursor), with human review on every change
- Architected real-time WebSocket infrastructure and event-driven systems, reducing data latency from 30+ seconds to real-time delivery
- Achieved a 90% performance improvement in React/Next.js applications through rendering optimization, code splitting and lazy loading
- Established testing culture from zero to 80% coverage with Jest and React Testing Library, cutting production bugs by 60% and rollbacks by 45%
- Built a Storybook component library for shared UI, improving cross-team reuse and design consistency
- Helped migrate to an Nx Monorepo with micro-frontends, reducing build times by 40% and enabling independent deployments across 5+ modules
- React
- Next.js
- TypeScript
- Node.js
- WebSockets
- Claude Code
- Nx
- Jest
- Storybook
Product Engineer
September 2024
- October 2026
- Built Mentoma, an Academy Operating System (Next.js, NestJS, Prisma/PostgreSQL): a manager can set up a branded site, add teachers and run a first class alone
- Made tenant isolation a guarantee, not a UI rule: scope is enforced in the database and in server-side role checks, with isolation tests before each release
- Built idempotent payments that can't double-pay: callbacks are idempotent, audit-logged and reconcile; a 14-day trial leads into paid plans, with plan limits enforced in code
- Built the student learning record: quizzes, assignments, feedback and attendance kept per student over time
- Shipped secure HLS-encrypted video and live classes, with Skyroom rooms created automatically and time-gated join links
- NestJS
- Prisma
- PostgreSQL
- Next.js
- Multi-tenant SaaS
- RBAC
- Payments
Frontend Engineer
March 2020
- October 2021
- Established code review standards and testing best practices, reducing production incidents by 70%
- Drove adoption of TDD with Jest across the team, reducing post-release bugs by 60% and improving user satisfaction by 35%
- Delivered optimized React dashboards with modular architecture, reducing page load times by 50%
- Set up a Storybook workflow documenting reusable UI states and speeding up cross-team review
- React
- Next.js
- TDD
- Jest
- Storybook
Full-Stack Developer
April 2017
- March 2020
- Owned end-to-end development of 5 complete LMS products, from database schema to React/Next.js frontend, serving thousands of concurrent students
- Achieved 75–90% performance improvements across mobile and desktop through rendering optimization and query tuning
- Built real-time systems with Node.js, Express.js and WebSockets for live quizzes and chatbot features, handling 10K+ concurrent interactions
- Designed SQL (PostgreSQL) and NoSQL (MongoDB) data models, improving query performance by 75% through indexing
- React
- Next.js
- Node.js
- PostgreSQL
- MongoDB
- WebSockets
Full-Stack Developer
September 2016
- April 2017
- Built a high-performance Next.js ticketing application for peak-event traffic, cutting page load time by 50% through SSR and caching
- Improved development workflow and CI/CD tooling, increasing team delivery velocity by 40%
- Next.js
- Node.js
- SSR
- CI/CD
Full-Stack Developer
March 2016
- March 2017
- Designed RESTful APIs, WebSocket connections and database schemas (SQL + MongoDB) for scalable learning management systems
- Architected and delivered 3 complete web products from scratch, migrating legacy PHP systems to Node.js/Express.js
- Node.js
- Express
- MongoDB
- MySQL
- REST
- PHP
