Development

Full-Stack

Built a multi-tenant SaaS CRM platform with WhatsApp integration, featuring a real-time live chat system (Redis Pub/Sub + SSE), AI-powered product/FAQ embeddings (pgvector), and a multi-app Turborepo monorepo (marketing site, dashboard, REST API)
Implemented Firebase authentication, row-level multi-tenancy, background job processing (BullMQ), RTL/Hebrew UI support, and an e-commerce settings module with vector search

Technologies & tools

Next.js

NestJS

PostgreSQL

Redis

Firebase

TypeScript

AI - RAG Systems

Designed and built retrieval-augmented generation (RAG) pipelines for domain-specific knowledge bases
Implemented vector embeddings and semantic search using pgvector for accurate context retrieval
Optimized chunking strategies, embedding models, and prompt engineering for production accuracy

Technologies & tools

Python

pgvector

OpenAI

LangChain

PostgreSQL

TypeScript

N8N Workflow Automation

Designed and deployed complex multi-step automation workflows for real business clients using N8N
Integrated CRMs, messaging platforms, databases, and third-party APIs into unified automated pipelines
Built custom N8N nodes and webhook-driven flows to handle lead routing, notifications, and data sync
Reduced manual workload significantly across client operations through end-to-end process automation

Technologies & tools

N8N

REST APIs

Webhooks

PostgreSQL

WhatsApp API

Node.js