I build fast, reliable product experiences with React, Next.js and TypeScript — from high-scale marketplaces to conversational search systems powered by NestJS, OpenAI and Qdrant.
| 4+ years | 300K+ users | 40K+ listings | 92+ Lighthouse |
|---|---|---|---|
| Product engineering | Marketplace reach | Live catalogue | Web performance |
Frontend
Backend, AI & data
Quality & delivery
Conversational product search that translates natural-language requests into useful catalogue results.
- Designed a RAG search pipeline using OpenAI embeddings, Qdrant vector retrieval and NestJS microservices.
- Combined semantic retrieval with structured SQL filters for budget, category, brand and use case.
- Added Redis-backed memory for multi-turn conversations and automated coverage for critical workflows.
Customer-facing commerce journeys used across a marketplace serving 300K+ active users and 40K+ live listings.
- Built responsive discovery, cart, checkout, payment and order experiences.
- Created reusable server/client and desktop/mobile modules with Next.js and TypeScript.
- Reached a 92+ Google Lighthouse score through rendering, code-splitting and client-boundary improvements.
Production source code is private due to client and company confidentiality. The outcomes and architecture above describe my direct engineering contributions.
- Product first: connect technical decisions to a clear user or business outcome.
- Performance by design: keep rendering boundaries, data flow and bundle cost intentional.
- Maintainable systems: prefer typed contracts, reusable modules and focused abstractions.
- Confidence in delivery: test critical behavior and automate repeatable checks.
Interested in frontend architecture, commerce products or applied AI?
Let's connect on LinkedIn or send me an email.